Hacker News new | past | comments | ask | show | jobs | submit login

I love your attitude. I have the exact same one. I'm so excited that another form of busywork has been eliminated from school by robots. Eventually they're going to start having to structure classrooms around learning instead of worksheets.



I really don't understand why essays are bad. Is there any kind of research that goes into this?

I agree with you that learning should be the goal. And any busy work that doesn't help should be eliminated. But I just don't know if we know what the right structure should be and if we can say for sure that things like writing essays don't actually help students cultivate their writing and critical thinking skills.


Essays are wonderful. It gives me great joy to write essays today, although they're sometimes called "blog posts" or "rants" or "emails" or "memos".

I hated writing essays in school, because the assignment was always "reproduce a work of writing that adheres to the arbitrary standards of the institution for grading purposes". Great writing can't be graded, as its value is entirely subjective.

As an example, here's an assignment that I might have completed under duress, vs. one that I'd complete voluntarily for fun:

"Explain how the theme of Chaos is expressed in Slaughterhouse Five. Use at least five supporting examples from the text and cite your references MLA style. Four pages minimum."

"Convince your best friend that Slaughterhouse Five is a terrible novel. Cite the text any way you please, ideally by comparing it to a book you think is actually good."

Whatever I produced for the first prompt can be graded by ticking off boxes and looking at my grammar; whatever I produced for the second prompt would need a thorough investigation of my own writing style and a framework of grading that takes into account my own voice as an author.

(To be clear--I don't think that giving my prompt in a modern classroom would immediately inspire students. They are far too burdened by the entire system for a single change to fix their experience. I am merely discussing the difference between "pointless essays" and "essays that authors care about".)


> Whatever I produced for the first prompt can be graded by ticking off boxes and looking at my grammar; whatever I produced for the second prompt would need a thorough investigation of my own writing style and a framework of grading that takes into account my own voice as an author.

The two prompts motivate the writer to practice two completely different skillsets; they're really not comparable.

The first one is focused more on the fundamentals of close reading and analysis. The writer needs to understand what the theme of "chaos" means, then closely read the novel or review their notes to identify literary devices or techniques that theme, and then tie it together in a "report". It requires the assignee to practice very basic skills... it's technical practice, not artistry.

The second prompt is the artistry - it's an assignment in discourse or rhetoric. The thing is, it's not possible to successfully execute the second prompt unless you've mastered the techniques from the first prompt. Beyond constructing logical or emotional arguments that may be tailored to your audience (your best friend), you still have to collect evidence from the novel. It might not be a list of literary devices, but if one of your arguments was that the book was poorly and confusingly written, you would still probably need to collect evidence of specific passages that support your claim. The whole point of the first prompt is to build the skill to do this, but with some hand-holding/constraints for practice.

I won't defend page limits, but even the reference style mandate is important because it has implications for how you actually write the essay. I deal with technical stakeholders all the time, and the amount of time that we could clear up issues if someone would just properly cite a reference can be ridiculous... perhaps those stakeholders were the teenagers who didn't bother to follow the citation guidelines for their literature class?


This really got me thinking, so here's another comparison to draw on:

As a teenage musician, I hated drilling my scales and etudes. Why bother when practice was limited and I had cool ensemble and solo rep to learn? What I didn't understand and appreciate at the time is that all the technical drudgery serves a very real purpose. Most of the existing pedagogy is directly pulled from, based on, or references real repertoire which you'll undoubtedly encounter in your musical career.

All those scales in intervals? Well, you can't even begin to make a complex passage musical if you can't execute the technique! Arpeggios in weird fingering/shifting patterns? Turns out that some very exposed orchestral passage necessitates that you use an oddball fingering because it's just not practical to do anything else in context. That entire development section in the concerto you need to cram for an audition? Good thing that one of your etudes book was effectively variations and embellishments on that section, so you can lean on muscle memory and focus on making it sound nice!

Essay writing is much the same. No matter what I'm writing - an e-mail, a project proposal, a performance review, whatever - I'm trying to communicate a point. That means constructing an argument and supplying evidence. And doing so in a way that your audience will grok without any additional intervention. You build this skill by practicing, sometimes in ways that seem dumb, boring, and disconnected from reality. Not every pedagogy is ground so well in reality as my music example, but I can't imagine that the cynical take that it's all purely to automate grading is a rational take on things.


I've jumped back into doodling on guitar after a mid-20's post band break up and... it's totally different this time. I am so much more interested in scales, building chords by manually and so forth, when I was younger I wanted to play songs.

The opposite is true for my math. I enjoyed algebra as a kid and hated trig and calculus. Now I am much more interested in calculus and don't like algebra algebra.

Sometimes there's different ways to learning, I have no idea.


> That means constructing an argument and supplying evidence.

Ah wouldn't it be fantastic if school essays were more like proving to your boss that you followed the spec to the letter and less like... Following the spec to the letter.

That's the difference and it makes all the difference.


What about convincing your colleagues that something about the spec is wrong?

It's worth noting that the same skills the "version 1" essay is supposed to teach should be helpful if all you need to do is compile a checklist and save yourself the hassle of argument.


> The first one is focused more on the fundamentals of close reading and analysis. The writer needs to understand what the theme of "chaos" means ...

I think what ends up happening in reality, at least, in my experience, is that you Google "Slaughterhouse Five chaos" and trawl the first several pages of results looking for information you can essentially copy+paste into your essay (with slight adjustments to get around automatic plagiarism scanners, of course).

I did still demonstrate some kind of skill, maybe research and the ability to condense information from many sources down into a single piece of work, but those weren't the skills you mention, and it was definitely not what the teacher was intending for me to do.

The second prompt the person you responded to runs into the same issues (I can Google "Slaughterhouse Five reviews"), but at the very least probably feels like a more engaging and compelling essay prompt to the student.


You're generalizing in a way that probably suits you and people like you. Not everyone though.

The second prompt would have sent me spinning, panic, want to run.

The first prompt, while being 'technical' and not what a future 'writer' would like to do at that point can be somewhat mechanically achieved and while I still wouldn't have liked it, I would begrudgingly do it and it probably helped me overall. It mentions using certain 'techniques' you would've learned about in class. I can apply that. They want a specific number of pages at minimum so that I don't just write 5 sentences to cover the 5 examples, sure, whatever.

Like learning math. You gotta learn the basics, learn the multiplication tables by heart. Do the same "compute (-7^2*13-7)+5/5" style exercises over and over. It teaches attention to detail and memorizing and following simple rules. If you can't do that it is very unlikely that a "closer to reality" question that someone that will later go on to become a mathematician would like working on instead would not send you into panic mode.


I disagree on the ability to execute the second prompt.

You would like them to execute the second prompt in a way that demonstrates the skills that the first prompt calls for. They won't. They'll just take the second prompt, and communicate exactly the same way that they already do to their friends, with similar skills and language. The result may be persuasive - particularly to their friends - but it won't develop analytic skills.


This is a classic pitfall faced by novice teachers. I fell for it too.

Assign the second prompt, and I guarantee you’ll get something like this as a submission:

“Bro, the novel sucks. Trust me.”

You can’t even give this a bad grade, based on the prompt. You can't say it’s not convincing, because they’ll say “you’re not my friend, this would convince my friend”

You can’t say it’s too short, because they’ll say you didn’t provide a minimum.

You can’t say it didn’t cite the novel, because you said to do whatever.

You can’t say it didn’t compare to other literature, because you said “ideally”.

Lesson 1 of being a teacher: give the students an inch and they will take a mile.

Teaching students is not unlike programming computers, in that they both take instructions very literally. If you are vague with a computer program, you know ahead of time because the program doesn’t compile.

If you are vague with an assignment you don’t know until you get it back. The more vague the assignment, the wider the variety of submissions. If you don’t tell them the font face you get a cursive one. If you don’t tell them the font size you get huge and tiny. If you don’t tell them the margins you get wide and thin.

So even if you would personally make a good faith effort at this assignment, it’s really better for everyone to be specific and follow the same format.


I covered this point in my last paragraph. The problem with students isn't that they can't follow directions or collaborate; it's that the hate school and don't want to be there.

One of the reasons they hate school and don't want to be there is that they are compelled to do pointless, grinding busywork, all day, every day. That's why they're using GPT-3 to fake their essays. Even three hours of reprieve from the system is worth cheating and dishonesty, and all the better if it helps their GPA.

I harbor no beliefs that a teacher can walk into the modern school system with a creative, exciting lesson plan and inspire students to perform. The system is broken and fundamentally flawed. It cannot be fixed. You are certainly correct that the best way to get consistent results out of your institutionalized students is to grade to a rigorous, clear format, but in doing so you've only played your part in reinforcing the exact system that drives them to cheat with GPT-3.


> The problem with students isn't that they can't follow directions or collaborate; it's that the hate school and don't want to be there.

I'd say it's a mix of both.

> That's why they're using GPT-3 to fake their essays.

I'm not actually sure about the motivation for most students. For students who I've caught using copilot on assignments, it's not because of the reasons you cite. Maybe it is for others.

> It cannot be fixed.

The main improvement that would fix most of this is to have higher teacher to student ratios. That alone would be a massive improvement, because then teachers would have time to engage students at a different level of attention.


Although the general gist of what you're replying to certainly evokes a response in me, I was not going to reply until I read your response and agreed with it's point.

Having taught writing in universities over a six-year-long stint, my experience agrees with yours.

Pragmatically, the reasons the assignments are structured they way they are isn't because bad faith by instructors, but rather because of the needs of students.

I don't blame the students-- they have a lot of shit going on.

At the same time, you're absolutely correct that making assignments in the general form we see them has more to do with what students actively demand: they absolutely do not want the kind of assignment suggested by the GP because anything other than a list of boxes to check causes profound anxiety in students.

Our comment threads here are excellent examples of what short writing prompts and assessments could look like, and I've gotten invaluable feedback on my writing from participating in internet threads. In this form of writing, there are distinct grades in the form of karma. And there are real stakes for communication, as I can easily fail to get my points across or even upset people. I even sometimes get useful responses that improve my understanding of the world or some topic.

As useful as that practice can be, if I had my academic advancement tied to these prompts it would cause me a great deal of stress: how the hell do I know in which contexts someone will read any given post?

In the context of the general internet I have a lot of easy ways out. I don't have to listen to dumb people, the poorly informed, or malicious trolls.

In the context of a classroom, I can't just tell the teacher "that's just, like, your opinion, man" because they are going to write down a letter and that's going to make my life easier or harder.

I'm not a big fan of contemporary education for reasons I could develop in book-length diatribes (I quite a PhD during my dissertation), but I get where students are coming from when they demand some clarity on how they are being assessed.


At the high school level it's even worse than you're describing. My daughter was taught using the Jane Schaffer method: https://en.wikipedia.org/wiki/Schaffer_method

Note the section on Body paragraph structure -- that doesn't begin to cover how structured the resulting essays were. I can still remember my daughter sing-songing "T, CD CM CM, CD CM CM, CD CM CM, SC." Every essay had to follow that exact pattern. Every commentary sentence had to start with the approved list of words, and likewise the concluding sentence.

It's entirely about how easy it is for the teacher to grade, and has nothing to do with teaching students to actually write. It was awful, and I did everything I could, including contacting the Jane Shaffer people, to criticize it and push back.


Good lord, that is even worse than I'm describing.

A great many people cannot fathom the fundamental truth that the majority of learning is useful simply because it allows us to enjoy to process of living in society for 70-90 years and for no other reason. There isn't any greater purpose to being "good at writing" other than "it makes communication and competition more accessible and convenient", but even that is a subjective value judgement. There are many sub-cultures who exist even inside our own that are perfectly happy being sub-literate.

A lot of people are saying "if kids don't learn dry, boring technical writing, then how will they write work emails and documentation?" The implication there is that if you can't write proper emails and documentation, you'll fail at your career. If the emails and documentation I receive are at all representative of greater industry (and I've worked in both aerospace and clinical research) then I can assure everyone that few people can write at a useful level and not only are they still employed but their companies are still around.

Given that no one can write anyway, why do we cling to forcing children through painful, humiliating exercises in standardization?


These essays are usually assigned in college-prep courses with the intention that the students will go to college where they will need to read/write in APA, MLA or other rigid, clear styles that would allow them to read/publish in specific academic journals. It's a high standard for the high-achieving students.


> Great writing can't be graded, as its value is entirely subjective.

Well... I have a lot of problems with this, on both sides of the fence.

Provocative start: How about we give up the idea that students are producing great work?

I mean, I imagine teachers think of the exercises they give as skills development.

One of the skills to develop is technical writing.

Surely beginning students do not know how to cite textual examples to back up their arguments. Understanding the role of evidence in making an argument should be fundamental to democracy. (Understanding that we are not living in that world recently in the US.) So they need to practice.

Choosing the theme is another skill. I don't have anything to say about it, but I don't have a problem with teachers asking students to try to figure something out before they write about it.


> How about we give up the idea that students are producing great work?

Almost all teachers are well ahead of you on this one. Far rarer is the belief that students are capable of great work. This is, in fact, my central point: if all of your assignments are bounded by the need for administrative convenience, creativity and originality cannot flourish.

Teaching writing through dry, separate "skills development" exercises is like teaching basketball as follows:

"Today we'll practice jumping from one ankle to the other. Today we'll practice reading a point guard centric offense. Today we'll be working on our vertical jump height. Now for the exam: demonstrate a cut behind the center and a layup. Hmm, your second step is slow, you get a C."

I am arguing that if you want to teach basketball, your students need to play a lot of basketball. Exercises will only really help them once they've experienced the game and have a burning internal desire to compete.


> I am arguing that if you want to teach basketball, your students need to play a lot of basketball. Exercises will only really help them once they've experienced the game and have a burning internal desire to compete.

Well. I played basketball since 7th grade. Not just played, trained 3 days a week. Before you can really play you have to master certain elements, otherwise it's just fooling around. And at first we trained all those elements separately. Balance, switching feet, turning. Just turning without the ball. Faster, slower. Jumping from left foot, from right. Catching the ball. Throwing it. Passing. Alone, with partner, against the wall. Hook shot, but just up, up, up, get the ball up. Hook from left, from right. Then hook standing directly below hoop. Left, right, left, again. Then adding movement. Over and over again. And then actually playing.


I play tennis, soccer, ultimate frisbee, jiu-jitsu, disc golf, rock climbing, hang gliding, whitewater kayaking, bicycling, and kite surfing, and I've taught many of the same to beginners.

I assure you that the instruction only works when the student wants to experience the final form, and they will not get any sense of what the final form is until they have "fooled around" and have an actual desire to learn the sport. At every level of their progression they need time to experience unstructured performance for fun. The same is true for writing.

Kids love to write when they're young and no one's hovering over their shoulders grading them. It's only once they get hit with the five-paragraph essays and the term papers and the dry grammar exercises that they learn to avoid writing, and associate it with boredom and stress.


Leaving aside the fact that your two examples assess a pretty different set of skills, one has to deal with the reality that every teacher has many students. It's simply not realistic to expect them to do "a thorough investigation" of every student's style and "voice." Imagine you have 50 students across two classes, each turning in one of these essays. How long are you going to spend on each one? 5 minutes? That's 4 hours of grading time. 15 minutes? 12 hours. Now consider that most of your school day is already occupied with teaching, prepping for classes, office hours, and other responsibilities.


I'll go further and ask "what if their 'voice' is just bad"? Just because I have a 'voice' doesn't mean it's necessarily good. Teaching students to be able to switch voices - some voices/styles are more appropriate for some types of communications than others. Recognizing these types, and being able to switch... that seems like it would be a thing to teach/learn. My early schooling was a long time ago, and I don't think I had the language to categorize all of this at that time, but I do have memories of doing this sort of stuff (mostly grades 6-8 where I had the same writing teacher, but later with various classes through grade 12).


I haven't been a teacher, but I have been a TA (while a grad student with a full course load) with a substantial weekly grading burden (e.g. 30 students' problem sets a week, with 10-15 problems each, for a class introducing concepts like formal proofs and basic number theory).

So I appreciate your point.

But I also remember being a student forced to churn out mindless formulaic essays with length and structure requirements. I hated it. I never liked writing until I finally had one good English teacher in high school who assigned and graded in the way you say is infeasible.

If a teacher doesn't have the bandwidth/capacity/skill/etc. to teach English well, maybe they should find something else to do instead of torturing students with mind-numbing assignments.


> 50 students across two classes, each turning in one of these essays. How long are you going to spend on each one? 5 minutes? That's 4 hours of grading time. 15 minutes? 12 hours.

Teachers with fifty students shouldn’t be assigning essays. There is no way for them to read them, which means they’ll grade by scanning for key words. That destroys the pedagogical value of an essay, this post’s point.


Sounds like the problem isn't so much with the assignment, but with the idea that assignments only have value if they are graded.


> the problem isn't so much with the assignment, but with the idea that assignments only have value if they are graded

There's grading and evaluating. Writing something you know won't be read, except for the purpose of being scolded for missing key words, is close to useless pedagogically. Someone motivated enough to learn from that (a) didn't need the assignment and (b) deserves better.


This! People learn differently - which IMHO schools don't usually account for - but I personally always learned best when putting something to paper (well, preferably the keyboard).

At my university, assignments were primarily used for guided learning - most of the grade came from the exam. If you cheat on the learning, you either don't _need_ to learn to pass the exam (meaning you should have a way to fast track), or you're asking to fail the exam, which hurts no one but yourself.

Maybe it's different in other schools? Cause I don't fully get the "Good." argument based on my experience. YMMV.


Sure— It's not at all tenable right now for teachers to provide in-depth critique on long essay assignments— that doesn't make critiques with avoiding, it makes long essay assignments worth avoiding.

I took a very difficult gatekeeper exposition class at a famously rigorous university a few years ago and loved it. We had to write a ton, but I didn't mind it because when you're learning to write, you need to write a ton. And boy did we. But not all classes there were like that! Some classes, mostly classes about writing were deemed "writing intensive," but others would require little more than a few pages here and there. The standard for that scant output extremely high and the intellectual critique was often blistering; the teacher concentrated on the subject matter instead of combing 50 paragraphs for split infinitive.

Currently, I attend a significantly less rigorous university as a full-time undergrad. I have 5 classes, including an elective on the history of a particular art form. The final will be a 10 page paper and 20 minute presentation preceded by a 2 page proposal. While this class requires significantly less written output than the exposition class, the assignment will still take an disproportionate amount of my time. The teacher has many students and no TA, so each paper will receive a cursory intellectual critique, but primarily graded on format and grammar. I'll not likely have learned more than if I'd written a really tight 2/3 page paper that got several serious critiques along the way.


All acts of deliberate writing can be examined for clarity, concision, fulfillment of their own purpose, etc. If the purpose of an essay is to teach writing, then the writing should be inspired and flow freely. It's only after someone can read and write competently that it's important for them to learn mold their writing for specific, dry purposes.

The second half of your argument is incredibly common, although I don't begrudge you for making it. Yes, it's true that teaching effectively and creatively is near-impossible given the current setup and demands of the modern education system. This should tell most people something about the worthiness of the modern system, but instead most of them defend it.


Sounds like a lot of data to process... Perhaps we should use gpt3 to grade the essays, too.


In that case, we can further automate this process and just leave humans out of it - pipe the student's generated essay back in, emit grade, done.

This reminds me of a moment that has stuck with me for a long time. Some time in the early 00s, I was wondering around town with a friend fairly late at night. We watched a waste truck picking up outside a building, there were stacks of Yellow Pages piled up, as they had just been delivered everywhere, like they used to.

My friend and I joked that they could have saved on transport and fuel by backing up the recycling trucks directly to the printing presses.


Reminds me of phone trees with a robotic voice telling to to press 1 for X, 2 for Y etc. When Google announced their automated phone tree handling, I thought, it's just two robots talking to each other over an imperfect medium, human language. Why not just connect the two systems together via API or something?


Those type of assignments are part of what "inspired" my daughter to skip out on all her high school English homework, forget the humanities, and study engineering in college. On the one hand, great, we need more women in engineering, and she's good at it (and graduated from a top college). On the other hand, the humanities are great too and it would be nice not to turn people off to them with that type of mind-numbing work that is not only subjective in content but also subject to the desires of the professor.


7th graders are highly unlikely to produce writing of such a great quality that it "cannot be graded". Maybe when Hemingway was in the seventh grade. And we can grade subjective things all the time. You can grade code based on something other than whether it runs efficiently, for instance.

Meanwhile, your prompt sounds like hell. And is far more subjective than the previous one.

> Convince your best friend that Slaughterhouse Five is a terrible novel.

So, before you even start the assignment, you let the teachers dictate the position that someone has? And you're not going to teach students to assess themes in books, so what will they judge Slaughterhouse Five on, aesthetics?

> Cite the text any way you please

Why on earth would you change that requirement. "Cite the text using method X" is a direct analog to "Coding standards dictate this naming convention". I would fire a "free thinker" who refused to adhere to the, sometimes arbitrary, standards for communication with the rest of group. Standards are good.

> ideally by comparing it to a book you think is actually good."

As a rule, I don't think convincing someone that a work of art is "terrible" should be done by comparing it to something else you "think is good".

> [No page limit]

You can trust an adult with that, but a seventh grader? Usually they need a page limit to encourage them to write more.

It seems like the following a complete essay that you would have to grade very well: "Slaughterhouse Five's lack of elves makes it terrible, because fantasy novels are just better and books like the Lord of the Rings have elves which makes it a good book [Source - My conversation with Johnny yesterday]"

Grading that well would be bad because it's horrible in every way.


As a teacher of English as a second language, I find the Jane Schaffer method quite helpful in teaching structure, idea generation, and the skill of supporting statements by providing examples. https://en.wikipedia.org/wiki/Schaffer_method

Of course all essays shouldn't be written exactly this way, but for students just learning about writing long form, it is a brilliant stepping stone to get the basics down. Too many students—and readers—can't differentiate between a "concrete detail" or statement of fact, and "commentary material" or statement of opinion. This method helps to distinguish them.


When challenged on why we had to follow rules that real writers often ignored, such as those "arbitrary standards" you hate, he responded: "They can break the rules because they already know them by heart. You need to learn them." That stuck with me. I firmly believe that writing essays about boring drudgery is a necessity to develop the skill required to write essays about what's important to you.

School is not about doing great work. It's about learning the tools which you can use to do great work. School does not ask you to do novel research, and so it doesn't ask you to write novel essays.


> School is not about doing great work. It's about learning the tools which you can use to do great work.

We have no compatible ideas if you insist upon this being the case. School is not "training for real life", it is real life.

The thought that it's appropriate and desirable to monopolize the bulk of the time, mental health, and attention of young people in order to keep them from attempting anything of consequence while they "prepare to do more work later" is deleterious to society.

(It is also not true that real writers know the rules by heart and choose to break them for their own effect. That implies that there is a real set of rules that people agree on and that every great writer is capable of producing a standardized set of writing that follows these rules. In fact, most great writers take great pains to tell personal stories of failure in school due to an inability and unwillingness to comply with their teachers, and this has been true since antiquity.)


I think you had a different experience. My experience with essays most typically was to choose your topic, research and write an essay. The standardized tests did usually ask you to choose from one of three topics, none of which was usually something you were enamored with, but that did force you to change from your comfort zone and adapt to situational necessity.


My favorite were "reproduce a report on these specific historical facts without plagiarizing the textbook," and before the internet was readily available. More a creative writing course than social studies.


Your prompt doesn't seem to actually allow for any critical thinking development or give any guidance to how you can more convincingly express your points or arguments. It also assumes you didn't enjoy the book, which I'm sure some students would have actually enjoyed, so now your prompt is even asking them to argue in bad faith, that doesn't seem to be something we want to foster honestly.

The first prompt requires the reader to critically analyze the book, by first requiring them to give it a charitable interpretation.

It is said that you cannot disagree with someone if you're unable to explain their position yourself in a clear and definitive manner. Obviously, what are you disagreeing with if you don't even understand what's the ideas behind the thing you disagree with.

That's what the first prompt would be about teaching you, to be able to understand other people's ideas and concepts, to look past your initial judgements and bias, give it a charitable interpretation, demonstrate you understood all this by summarizing the idea in a 4 page essay of your own, with supporting references to tie it back to the source, showing the source does in fact argue for these itself.

Once you can do that, you have gained the right to go on with your own disagreement and write that essay, which would be your second prompt. Though honestly, your second prompt seems to be geared more towards discussing the entertainment aspect of the book, and not the ideas and concepts it contains, so again it's not that much about critical thinking, because critically there's little to argue about a "I prefer the color red over blue."

Personally I think you were trying to get at something else, maybe your point was just, come up with assignments students enjoy and can have fun with?

I think this is always true, but some things are just boring to some students, maybe you just don't enjoy reading, writing or even critical thinking, or any of that stuff. I don't know if there's much you can do in that situation. Maybe the solution is more fluid classes, let people move at their own pace, pick their own areas of interests, even if that's directly going to a trade, skipping on literature entirely, etc.


What exactly do you think "critical thinking" is?

The first prompt presumes to pick what was important about the book, mandate the form by which the book will be analyzed, and set up a minimum amount of effort before the writer can quit.

The second prompt picks a very general bit of opinion and then demands an open-ended argument requiring original thought. In fact, it's even better if the reader liked the book, because it forces them to write as if they didn't, and opens them to the possibility of a satirical essay.

(Note that I am aware that an average modern student wouldn't like the second prompt any more than the first, but that has to more with the system than the prompt. I'm speaking about the pure act of teaching an interesting writer to write well.)

Critical thinking requires both the desire and ability to think outside of frameworks that were predetermined by authority. This is part of the reason that modern schools are so bad at "teaching critical thinking skills". The most basic form of critical thinking, in fact the first openly critical thought that students have about learning--"this is a waste of my time"--is suppressed for the convenience of the administration.

> Maybe the solution is more fluid classes, let people move at their own pace, pick their own areas of interests, even if that's directly going to a trade, etc.

I could not agree with this more. I believe that 'school' should be life-long, year-round, and optional. Ideally we'd go in and out of some type of formal education until we died. However, this level of societal flexibility is directly incompatible with modern school.


I'll go a step further and defend "busywork". Unfortunately we don't live in a perfectly efficient society so being able to learn and execute a formulaic task that might not seem valuable is a necessary skill for most adults. Especially since individual actors may not have all the necessary information to fully assess the value of a task within a larger framework.


It's an interesting question. I'd personally love to see more effort being given to "defend your point as succinctly as possible." I wrote a lot of "minimum word count essays and in the "real world" clarity trumps length every day.

Length is used as some proxy for rigor, but we know it's only a proxy.


I like this idea. Fwiw I feel I learned / grew a lot more in one technical writing class in college than I did in all my years of English classes. Later I picked ip a book “writing with style” that was also quite helpful.

One issue for me in English was I really was not interested in the kinds of essays the English teachers were interested in having us write, eg coming up with a thesis on plot themes in Shakespeare. Just not my thing and so I couldn’t get anywhere in those classes.


I found that in elementary or secondary school page minimums were used to as a cheap proxy for effort. In college page maximums started appearing to encourage concision. Different needs for students at different points in their education.


My high school English teacher adamantly refused to read any essay over the length of one page, double-spaced. I learned more from that class than any other writing class I took.


+100

When I write today, for work, the challenge is always to write less. My VP might have time to read two pages; he's almost certainly not going to read anything longer.


I learned from one of the comments on my original post that many scholars have been saying this for a while, and that there's in fact a book that makes the same point!

Why They Can't Write: Killing the Five-Paragraph Essay and Other Necessities https://www.amazon.com/Why-They-Cant-Write-Five-Paragraph/dp...


The American educational-system essay is an artificial construct designed to be easy to teach and easy to grade, without being useful for the purposes of educating the essay author or the audience or being persuasive.

Writing is generally good. Expressing yourself is good. Analysis and critique are good. The artificial essay is useless.


Especially the 5-paragraph essay. Doubly so where paragraphs are rigidly-defined as "a collection of five or more sentences". I argued with a lot of teachers over that one.

All it did was teach kids how to write boring simple sentences to meet the implicit "punctuation quota". An equal amount of content compacted into a couple of complex/compound sentences would actually result in marks off.


Which is exactly the type of writing that an AI excels at. Educators have made their own bed, and GPT-3 is just tucking them in.


The 5-paragraph essay is the larval form of the 6-page research paper. As a scientist, you will similarly face reprimands from reviewers if you deviate from the format.


Is it better to teach a good writer a specific format, or to teach a specific format in order to teach people how to write?

Most people will not become scientists. Those who do can quickly pick up the additional burden of formatting requirements, when necessary.


Most people will encounter venues where they need to conform to arbitrary and seemingly pointless formats in order to publish there. Becoming comfortable doing so without throwing a fit is a skill that needs to be learned to operate in modern society.


I'm not throwing a fit. I'm saying that the five-paragraph essay is supposed to teach clear writing, and it doesn't.

If you want to teach conformity to arbitrary and seemingly pointless strictures, the rest of the public education system does that already.


I didn't mean to say it's you throwing a fit, I meant to say that students are throwing fits. Unless you're a student? Sorry I didn't convey that clearly.

I wouldn't say that the purpose of the 5-paragaph essay is to teach clear writing. It serves the same purpose as mandating a format for a conference; no format mandate means you get N formats, which makes evaluating them much harder.


You have to learn the rules before you break the rules. The 5 paragraph essay taught me a lot about how to create a convincing written argument. The rules are:

1. Start with a hook. Engage the reader.

2. Present the ideas (2-3) that you will be discussing in the essay. Have you ever read a rambling blog post? Yeah, they suck. There's no direction and you don't retain much.

Also, 2-3 ideas in one essay is a great number. There's all those studies that say we can only hold 5 things in working memory at once, blah blah blah. Keeping the essay focused on a few core ideas helps the reader retain them better, and the writer to have a well defined scope.

3. Extrapolate the concise ideas in 2-3 concise paragraphs. Ever read a rambling blog post? Yeah, they suck. Telling students to keep paragraphs in 3-5 sentences helps the essay communicate the ideas in a concise manner.

It also helps block out the text in small visually appealing blocks. Ever read super condensed very long paragraphs? Yeah, they suck. It helps to break up your thoughts with some whitespace. (It's almost like coding benefits from this as well...)

4. Conclude your essay. Reiterate what you wanted to cover. This helps the reader retain the ideas, and it allows the author to tie up the ideas in a nice bow. I love when I finish a book or essay and everything comes together and reaffirms what I've been reading the whole time.

This format is not only great building blocks, but it helps you write larger volumes. If you repeat these small steps several times, you create chapters. If you repeat these steps on a macro level, the chapters tie together into a cohesive piece of literature.

These unnecessary "quotas" may sound meaningless, but a lot of people have thought very hard about how to create basic building blocks writers can follow. These building blocks allow the writers to create concise, well formed arguments. "Boring simple sentences" are extremely conducive to clear and concise writing. I'll take boring sentences that form complex ideas over complex fluff that describes nothing any day. (This is almost analogous to good code design too, weird...)


I get the goal of the framework, and I did then as well. It just isn't expressed well, and it's even worse in execution. I think kids are smart enough to handle a more ambiguous "five-section essay". All the same rules apply, but marks off can then be directed towards lack of clarity or belaboring a point. Truly bad writing, y'know?

Maybe it's a matter of opinion, but I believe that the grading process is a sufficient guardrail. When the theories are made rigid, it's usually just to ease the grading. I'm okay with that in the right context. In this instance, however, I think it's self-defeating.


When I taught a class where I assigned essays I confirmed my own suspicions from when I was assigned essays: the majority of the sentences committed to paper are awful drudgery. I then flipped the requirements on their head and removed minimum page requirements and instead had extremely strict maximum page requirements (with of course the objective material requirements of the essay itself). So much improvement for everyone involved. I had to read through so much less pointless material and the students were forced to focus their ideas in a succinct way to be able to get to all of the objectives of the essay in the limited space. Everyone saved time and did better.


When I taught a class, I had a strict 3 page maximum on lab reports for this reason. So many people are used to turning in 10's of pages of drudgery. I just want a few that get to the heart of the issue.


Being a concert violinist is fun but playing scales isn't fun. I can't wait for the robots to learn how to play scales for me so I can focus on becoming a concert violinist.

;-)


Writing essays is not the problem, grading the essays and grades in general are the problem.

Feedback is obviously invaluable, but the point of grades, as used today, is solely to gatekeep who are allowed access to the next level of education. So instead of constructive feedback, it has become a set of filters entirely divorced from actual learning.

(And yes, I got good grades, I just hate that so many people I know were denied opportunities based on a shitty system, wildly not fit for purpose)


I think you’re going at it backwards. If you are going to ask a student to write an essay, it’s on you to be able to show that it’s the best way to help that particular student cultivate their writing and critical thinking skills.


Perhaps I'm just... thick, but if a goal is to have someone write, say, 8 pages of thinking and ideas about a topic... I'm not sure there's 'better' ways than to have the person write 8 pages of thinking and ideas about a topic.

If the goal is wrong, perhaps just don't do it, but... "it's on you to show that it's the best way". I don't get it.

If I want to see that a student has writing skills, I would think expecting them to write is somewhat definitional?

Maybe it's on someone else to 'show' a better way to demonstrate writing skills that doesn't involve writing.


> want to see that a student has writing skills

This is a fine goal. Having “someone write, say, 8 pages” is not, it’s a task, and a tedious one at that. No good writer starts with a page goal. It’s a common criticism by great writer’s of bad publishers.


Personally, my experience with page count assignments were... loose. The goal was never "write 8 pages" but "write about idea ABC" (book just read, subject we just studied). '5 pages' or 8 or whatever was a guideline, with the expectation that to get XYZ ideas across, it'll probably be around that length. If the guideline was X words or Y pages, and I got the ideas across in less (or more), but the ideas were strong, I still got a good grade. Perhaps some of that has changed, but... "write an essay about $foo", and you turn in 2 paragraphs... you'd get marked down.

Again, it's been awhile since I've been in middle/high school, so it may have changed some.


These days it’s often a word count rather than page count because page counts are too easy to game. There will be an upper and lower limit and when you submit the file, it might be rejected if your paper doesn’t meet the criteria.


If you're going to up-end hundreds of yers of educational theory and practice, it's on you to be able to show your work. Not all progress is forward.


Does that need to be shown? Seems tautological to me. You learn to write by writing.


The writing part isn't necessarily being questioned, but the form (essay) is. The tweet thread pointed to in the article has some pretty good suggestions.


The twitter thread is suggesting prompts for essays that an AI can't handle. So the form hasn't changed, they're just trying to stay ahead of the threat.


I was thinking like you until the day I realized that a lot of learning actually implies busywork, like being good at a sport implies a lot of busywork like exercise and nutrition.


Busywork in sports and nutrition are all in obvious service of a more important goal (increase performance in sports). There's a clear incentive and reward to your performance.

The same is not true for busywork in school. It is disconnected from life outside of school in a very fundamental way, in that there are no inherent rewards for being able to complete busywork--all of the rewards come from social cleverness, competitive instincts, networking, etc. If you behave like a good student in the professional world--keeping your head down, getting your work done on time, following all the rules--your friends who go out drinking with your boss will blow past you in your career, and you'll be pigeonholed as a drudge-work guy.

People treat school as if it's training for real life. But it is real life. Why is everyone so completely convinced that it's got to be boring, miserable, and institutionalized?


Most jobs in the real world don’t benefit from social cleverness. They don’t even offer the opportunity to schmooze with the boss. They are: “Drive this exact route and deliver boxes, within a time precision of plus or minus 30 seconds.” and “Repeat this physical motion on this part 7000 times while keeping yield over 99%.” and “Ensure TPS reports are filled out exactly using the correct template and are on Lumbergh’s desk every Friday morning 8:30.” I would argue the repetition, rule-following and drudgery of school adequately prepares workers for this adulthood.


First, I don't believe you're correct about social cleverness not helping drivers, line workers, or office drones. If anything, the more basic the work, the bigger the rewards for social cleverness are, as you use your relationship with your superiors, peers, and inferiors to influence the work day. (This could be becoming less true with fully metrics-driven establishments like Amazon warehouses.)

Second, is that society (where a huge number of people are consigned to that type of labor) a society that you're interested in continuing to build? If so, continue to support the methods used in modern school.


In order to do that, we would have to be able to measure “learning”, and so far, we have only been able to measure “ability to regurgitate facts”. It does not help that the mainstream educational paradigm is “to fill the blank slate that is the student”.

See: John Taylor Gatto’s book on this subject— The Underground History of American Education.

There are other educational paradigms, but they are radically different — no tests, no curriculum, no “fill the blank slate”, no measuring.


I knew I'd see Gatto come up eventually. I owe my life to that man. Before and after reading his work is like before and after the Wizard of Oz switches to color.


Gaining a skill is how we can measure learning.

Writing is a skill.

Writing clearly, grammatically, persuasively, and logically takes many skills.


That is measuring from the outside, and quantifying it. How do you measure “skill”? You still look at the results, rather than the skill that develops inside a person. It’s still the same problem as measuring “learning”.


Learning how to write is _really_ important.


Maybe, but the flip side is that I'm wondering how many schools will look at the opposite and try to start using machine learning to grade essays as well.


So we end up with machines grading machines, while pupils and teachers are left doing... something. Isn't it known fact that one cannot fix society with technology?


How do you want to learn good writing skills?


By generating text from AI, evaluating it, and piecing it together with other AI generated text.


LOL, thanks!




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: