Drawing on Malesic’s essays in The New York Times, The Atlantic, and The Chronicle of Higher Education, this thoughtful conversation examines how educators can foster intrinsic motivation and rebuild student confidence in reading. They discuss strategies for reintroducing long-form reading, the shifting value of writing assignments, and whether the academy can still champion intellectual growth beyond job training.

Eric Mazur (00:25)
Thank you for joining us today for this episode on reading and generative AI in the Social Learning Amplified podcast series. I'm your host, Eric Mazur, and our guest on the episode today is Jonathan Malesic John is a writer whose work has appeared in the New York Times, the Washington Post, the Atlantic, the Chronicle of Higher Education, The Guardian, and many other publications.

His writing has been recognized multiple times in the best American essays and in the best American food writing. His latest book, The End of Burnout by the University of California Press, was selected as best book of 2022 by Amazon and the Next Big Idea Club and is being translated into 10 languages. He's a former full-time theology professor who holds a PhD in religious studies from the University of Virginia.

He now teaches writing at Southern Methodist University. John, thank you for being here today.

Jonathan Malesic (01:24)
Thank you so much for having me, Eric. It's wonderful to be here.

Eric Mazur (01:25)
Well, it's wonderful.

Same here. I've read many of your recent writings on reading, writing, and chat GPT, you know, with much interest. And reading between the lines, I found that you, like myself, and I'm sure many of our listeners here today, are struggling to understand the repercussions of generative AI on learning, on reading, and on writing. I find my own thinking constantly evolving.

And you state in one of your pieces, and I'm quoting here, as much as students worry me, they also give me hope. As I talk to my own colleagues, I hear many concerns. I'd love to focus on the positive. So what gives you hope in this current age?

Jonathan Malesic (02:11)
Yeah, what gives me hope is that despite all of the incentives that students feel to, you know, maybe to cut corners on their studying or, you know, to, you know, to check out of discussion, to let their, you know, classmates take the lead. And obviously that's not a sustainable approach to a discussion class.

they have all of these incentives to avoid real learning. And yet, I see them do real learning every day. You know, it's not 100 % of students, it's not 100 % of the days, it's different students, you know, may kind of check in and out, but they do often defy those incentives. You know, the external incentives of‍ careers and other pressures. And, you know, there is this motivating force within students to want to learn.

Eric Mazur (03:15)
I think that's the basic innate human nature, right? It's part of self-determination theory, the three basic psychological needs that we have, growth, autonomy and a sense of community. And I recall a number of years ago I invited a speaker at Harvard on game design because I started to think, you know, maybe we can gamify learning in order to promote the intrinsic motivation that you were alluding to. That talk was eye-opening in the sense that this person who was from the Stanford D. School said, the basic principles that motivates people to continue to play a certain game is learning, right? In other words, you have to learn a skill to get to the next level. And game design, it's not that you use gaming to motivate learning. No, it's the other way around. You use learning to motivate gaming. That was such an eye-opening discovery for me. And you're saying something that actually rhymes completely with that. So in spite of that hope, I think that reading and writing are potentially both under attack. In fact, even before the advent of Chet GPT, Gallup reported that Americans read on average 25 % fewer books now, or actually that was two years ago, than they did just 10 years ago. And in an opinion piece you wrote this past October in the New York Times, you mentioned that you hadn't assigned an entire book in four years, although I think you've changed that now.

You know, that college professors nationwide have noticed a decline in students' willingness and ability to read. What do you think are the most significant factors contributing to this trend, which predate chat GPT?

Jonathan Malesic (05:06)
Yeah, absolutely does. When I was teaching theology full-time, I was primarily teaching general education courses, the kind of courses that students don't really want to take. Very few students are eager to take a required theology course. But I would require usually four or five, maybe six.

books and there was a semester or two when I required nine, mostly fairly short books. yeah, I mean, it's not that those students were, you know, super superheroes. They were ordinary students. And I knew that I knew I was pushing them a bit by requiring nine that one semester. But I had seen that, you know, with the right incentives and the right kind of prodding and cajoling, you know, they would, they could do it. You know, very few people are just dispositionally inclined to read a couple hundred page theology book, but they can do it. You know, we can, we can. Give these incentives and people will see that it was worth it Yeah, and so I think that I'm only talking about like a decade ago this was for me at least And so quite a bit has changed in that time because I mean I hear from colleagues who yeah, I mean who you know cannot Feel they can't assign whole books anymore that students just aren't up to the task and certainly one big thing is the pandemic disruption. That's, you know, so many students lost years of really, you know, good quality instruction in school and have struggled to recover even several years later. I think in addition, you know, we could say that the way we read has changed.

Yeah, I mean, I don't doubt the statistics on the number of books that people read, but people will actually read a lot today. You know, even if you spend a lot of time on TikTok, a lot of people, a lot of TikTok videos are made so that you don't have to have the sound down and you can just read the text. And we're constantly messaging people. You know, so we actually read quite a bit. It's the question of what, you know, are we reading these long narratives, long arguments, you know, that's a worthwhile thing to know how to do. So that skill may have atrophied a bit, but reading is still quite prevalent, I think.

Eric Mazur (07:57)
Right. Yeah, but it's the quality of the reading and the growth from that reading too. I guess that from texting each other, that's reading on a very different plane in a sense than reading a novel or literature. or even the news. So what role do you think Genitive AI plays in, not only in that, in their reading habits?

Jonathan Malesic (08:22)
Yeah, I would agree.

Yeah, I mean the big worry that I have and the thing that depresses me is the capability of uploading an article, even potentially an entire book, to Notebook LM and just getting a summary. you know, I mean I'd be lying if I said that the summaries were bad. I mean, they're not.

And, you know, if all we wanted was for students to get the handful of main takeaways from a book, then I guess, you know, there's no problem there. But that's not what we want. We want to cultivate other things and to understand, you know, like in a literature class in particular, you know, the nuances of language. We want students to exercise, you know, a kind of patience. You know, a novel is not the only kind of long text a student might read in their lives. You know, they might read very long technical reports where you won't want to just kind of hand it over to Notebook LM partly because it could be proprietary information that you don't want to hand over to a third party. But.

You know, those are abilities that students need and the trick is how to give them the incentives to build up those skills. And I haven't totally, I certainly haven't figured it out.

Eric Mazur (10:04)
Are you familiar at all with Sparousal? So, okay, you've used it. So, do you think that the, so what I found was interesting, you know, for a long while I resisted

Jonathan Malesic (10:06)
Yes. Yeah, I've used it. Yes.

Eric Mazur (10:16)
allowing video on perusal because to me video puts, you know, the student, the audience in an even more passive role than anything else. When you read, at least you have to engage your brain in imagining things, much less so when you listen to somebody. But then during the pandemic, I thought we have to allow video, so we permitted video. And to my surprise, students, you if you look at, if you upload a video to, let's say, you know, any other video platform, or whatever, where you can see the statistics. You find that students mostly play the video at two times the speed. They try to fly through it as fast as they can. In the beginning of the semester, they watch the whole video. And as the semester goes on, they watch less and less. To my surprise, when we allowed video on Perusall, the total time spent on watching the video was significantly more than the video itself because of the interaction that occurs while watching. the video, which gives also your brain time to think beyond the information that is just coming at you. So maybe the same thing is true for reading. If you provide an opportunity to exchange ideas with others, to ask questions, so maybe there's hope there.

Jonathan Malesic (11:31)
Yeah. And I mean, that's what I've seen. You know, I have not used Perusall in a couple of years. Not for any, you know, like real dissatisfaction with it, but you the pressures of what, know, the incentives that I have and the pressures I have as a writing teacher. There's just always so many different things to try to do. And yeah, so I haven't used it in, in the kind of AI era.

Eric Mazur (11:51)
Absolutely. Absolutely.

Jonathan Malesic (11:58)
But yeah, I mean, I agree. Like what I've seen it be good for is exactly what you're talking about. you know, though, I mean, that can also happen in just a well-managed discussion, though so often, you know, students may have quite a bit to say, but they're hesitant to voice it. And...

You yeah, I think that a benefit of perusal is, you know, giving, giving them an alternate forum.

Eric Mazur (12:25)
Going back to what you said earlier, you used to assign nine books and then you completely stopped and now you start assigning. One, I forgot exactly how many, but you started to bring back book assignments. So one is what are your hopes or your aspirations with that one assignment? And also I'm sure that there are many people who might be listening to our podcast here who are struggling with their own students' reading habits. So what are you hoping to achieve and what advice would you give to other educators?

Jonathan Malesic (12:35)
That's right.

Yeah, I mean, I'm hoping to build students' confidence more than anything. I think that probably, you know, well, I mean, not probably. mean, it's heartbreaking. mean, my students will sometimes tell me that they aren't confident readers, that they give up on reading very quickly because, you know, they may struggle to focus or they may struggle, you know, with the vocabulary or something. and I want to prove to them that they can do this. I want to show them, look, you know, here's this awesome reading. you know, let's, let's start out, you know, let's start it in class. let's read, you know, in during class time with me here by your side. When I do that, you know, they don't struggle so much. And my hope is to say, look, you can do this. You know, and so my hope is that to build that confidence, it'll become easier to assign more stuff outside of class. And then, you know, their gains in learning will just skyrocket from there.

Eric Mazur (14:08)
Do you think it's really a confidence problem or is it an attention problem?

Jonathan Malesic (14:13)
Students often describe it as an attention problem. That's for sure, and I'm sure that that's part of it. I know that when I...probably less when I'm reading than when I'm writing. If I encounter difficulty, I will very often just like habitually, you know, grab my phone and see if I got any email in the last five minutes. So I'll switch tasks because I've encountered some kind of obstacle. And, you know, I'm speaking, you know, purely from my own experience and not from any, you know, kind of scientific knowledge of this. But at least in my case, yeah, could be like encountering an obstacle may be an opportunity to let your attention stray for students.

Eric Mazur (14:56)
And there are many more opportunities to let the attention stray now than ever before.

Jonathan Malesic (15:08)
Absolutely. Yeah, yeah, absolutely.

Eric Mazur (15:20)
Now that you moved on to writing, which is actually I wanted to move from reading to writing, so this was very appropriate. In, I think it was last year,you wrote an essay for the Atlantic entitled What Chat GPT Can Teach My Writing Students. And in that piece, you argue that learning to write trains your imagination to construct the person who will read your words. So true. I mean, I wrote a textbook in physics. It took me 22 years to write it because I realized I see the world from the perspective of an expert, and I have to train myself to see it from the perspective of a novice reader. However, as an author, you, and I'm asking myself that same question, what's the point of writing if people read less and less?

Jonathan Malesic (16:05)
one thing that I encourage, I also encourage my students to do in the first year writing classes I teach. And this is another thing that they resist for reasons I get into. But is like you gotta use the fewest words possible to say what you wanna say. You have to respect your reader as someone who could be doing a million other things, but is taking the time with your words. And you should respect that person's time and attention, reward it. Make reading pleasurable and you know, make make their make make it worth their while that don't don't, you know, take more time than you really need. And so if a writer is doing that, then an AI summary of what they write will will necessarily lose a lot of important information and pleasure,

Eric Mazur (16:48)
Mm-hmm. as an instructor like you and as somebody who writes a lot professionally in my discipline, but I've also written a couple of books for students. I've been struggling, and I write tons and tons and tons of letters of recommendation. In fact, I have a pile with the December 1 deadline looming. I've been struggling wrapping my head around the following scenario. And maybe the scenario is a little bit far-fetched, but I think it helps us think about the future of human communication. Let's say I have to write a letter of recommendation for, which is the bane of my existence, for...a student in my class or a graduate student who is applying to a job. Now what you do is you just make a bullet point of things you want to highlight and you can feed it to ChatGBT to create a letter of recommendation. I haven't gotten that far yet, but I could do that and I've played with it and I know a pretty nice letter of recommendation will come out. You send that off and on the receiving end, the person takes your letter feeds it to the chat GPT asking to provide a list of bullet points. So here we are throwing away almost language as a way for human communication completely, and we take a bullet list, have a translation algorithm to something that we used to find the appropriate way of communicating, and then invert the process on the other end to get a bullet list back to save time. Is this where we're gonna head with reading and writing?

Jonathan Malesic (18:38)
I, that's certainly what I fear, but just in, your description of that scenario, it, makes me think, okay, well, this is maybe an opportunity to change this expectation of what a letter of recommendation should be. that perhaps the letter of recommendation does that it we imagine it's doing one thing, but really it's doing something else. we don't value the actual writing of it, well, maybe it's the wrong genre for the work that we're asking it to do. And so perhaps that needs to evolve somewhat. And yeah, to bring it back to the writing classroom, yeah, I mean, as many others have observed, The world does not need more papers on Plato's Allegory of the Cave written by undergraduate students across the United States and other countries. But it's worth it for the students to do that as an exercise. There's value in it for them, even if

Eric Mazur (19:50)
What is that value? I think I know where you're going. I presume that you mean the value is in developing a skill that might be relevant to them in their future careers. But I'm guessing here.

Jonathan Malesic (20:07)
Yeah, I mean, in their careers and beyond their careers, just in their ordinary lives outside of work, it can teach you to think clearly, to analyze, to explain something. sort of like I mentioned in that Atlantic essay, it's..if these kind of if you as the teacher set up the conditions well and I think this is hard to do I don't do it very well then you know the writing field can feel like the student is is communicating with a real person who really wants to know this is kind of the artificiality of the way that writing is taught in school you know I was telling my students the other day that It's only in this classroom where the person you're writing for absolutely must read what you're writing. I have to like any it's my job. I'm duty bound. If someone were to write me, you know, a boring or incoherent thing, you know, out in the world or encounter it, I can just go, well, this isn't any good and skip it.

So the students, there's no stakes in a way. Like the students don't actually have, I say, well, write an introduction that grabs your reader's attention, but their actual reader, the attention, my attention isn't at stake here. I have to read it regardless.

Eric Mazur (21:33)
Mm-hmm.

Talking about, I think you made an important point because I was talking about skills important for the career and you broadened that to the life in general. But you do mention in one of your writings, I forgot which one it is right now, the role of universities in promoting the idea, well it's true overall with US World News reports and so on, of universities promoting the idea that a degree is primarily about earning power.

Jonathan Malesic (21:38)
Mm-hmm.

Eric Mazur (22:00)
So what responsibility, if any, do universities bear for this trend? And what should institutions, faculty as a whole, so to speak, do to encourage students to value reading and intellectual pursuits?

Jonathan Malesic (22:19)
Yeah, I mean, I think that they have to demonstrate that there is value to intellectual pursuits apart from their ability to make you money. You know, that too is difficult because, you know, teaching really, we may as teachers embody the life of the mind in a way. It is also our job. So,

you know, we can say, you know, it's worth it to read, you know, just on your own and all of this, but then, well, how much do we actually do that apart from our jobs? And, you know, I just think about my own mentors as an undergraduate student. I mean, I, I just wanted to live a life like theirs, or at least like I imagined theirs what to be. they just seemed like interesting people. Yeah.

Eric Mazur (23:07)
As an academic, you mean, or as a writer.

Jonathan Malesic (23:10)
Yeah, I would say as as an academic, but I think even if I hadn't gone into academia, I would hope that I would have wanted to be, you know, some sort of, you know, well read reflective person. Because I think that that's, that's a worthwhile, that's a worthwhile, I don't know, disposition, maybe to, to have available to you, even if that's not who you are all the time.

Eric Mazur (23:43)
Right. And of course, you know, speaking of us as academics, we tend to sort of, I find that I often have to fight that for myself. We tend to project our own experiences onto the students in our classes, not realizing that in a sort of steady state world, each of us academics is entitled to one single successor, only one of our students can become like us, or soon the world will be covered in academics, which I don't think would be a very good thing. Going back to 2006, I saw an essay you wrote for the Chronicle of Higher Education about plagiarism. And I want to quote something you wrote in that article. A student who plagiarizes refuses to be educated.

there shouldn't be room in my classroom for that kind of a student. Indeed, person is not really a student at all. So now with Generative AI, has your view changed?

Jonathan Malesic (24:39)
I think that my view of plagiarism has changed a little bit since then, though I don't think generative AI really affects my overall view. I've softened a bit. That version of me in 2006 was... I mean, I was personally offended by plagiarism, and I encountered a lot of it. That was early in my full-time teaching career, and I now think about it in terms of, you know, how, what, what incentives do I, or does the college wittingly or unwittingly give students, you know, where perhaps have we failed, and are, you know, perhaps even encouraging plagiarism or other kinds of shortcuts. And, you know, I, I use it more often as kind of a learning opportunity. You know, it's like, okay, well, this student screwed up, maybe through ignorance, maybe not. And, you know, let's talk about it. And it's like, all right, let's do this again. There's something to be learned there. You know, they are a student, they're not doing their best as a student, but they can be turned around. But yeah, I mean, I think that

You know when when I was starting out Wikipedia and spark notes and things like that were the real sources and in a way it was it was I found it easy to catch plagiarism even before turn it in was invented Because I'd be like yeah that phrase doesn't sound like one of my undergraduates and so I'd Google it and there it is And we briefly had like the period when turn it in was pretty good at detecting it and I think now You know, we're back to a thing where it's it's just so hard to detect like what's AI generated and what isn't I don't think we have reliable detectors right now and yeah, I don't know. I don't know what to do.

Eric Mazur (26:37)
Well, I struggle with that too, but you you mentioned the word shortcuts, right? And I think that it's innate in human beings to find shortcuts. mean, if you look at all human advances over time, technology and so on, it's all about increasing efficiency and productivity in a sense, right? I mean, we're not going back to making stone tools each time we have to build something. And each time there's a new technology that appears, people are first very alarmed when the technology is being used. Think of the calculator, right? mean, initially it was forbidden for students to bring calculators to any type of exams. Now we don't even think about it anymore, and students can use calculators. And then the internet came around and you're not allowed to look anything up. And I think that more more people are discovering, giving people access to information is actually beneficial because that's how they're going to operate in later life. You just have to ask more relevant questions. Maybe the same is true with Genitive AI. And if somebody can be extremely creative using Genitive AI, that should not be seen as a negative, but as a positive. I don't know, where do you stand there?

Jonathan Malesic (27:48)
Yeah. Yeah, I'm not yet sold on that. But I certainly appreciate the argument.

I suspect I have no way to prove one way or another, you know, what I see in student papers in the last year or so is that they tend to be pretty polished on the sentence level. aren't a lot, there aren't the kind of grammatical mistakes that I might have seen a decade ago, which makes me think that even I'll just call like generative AI light. You know, things like Grammarly or something are being used pretty widely. You know, I think that there's a lot of value in learning those grammatical rules, even if you are going to rely on generative AI. But...

Yeah, is it the worst thing in the world if students find it easier to write, you know, polished sentences? I don't know. Maybe not. I'm not, I'm not certain.

Eric Mazur (28:52)
Ultimately, have to get the software to produce something. I completely understand your ambivalence, because I have the same ambivalence. But you know, in the sciences, we have a lot of graduate students from China whose writing skills in English are sometimes very limited. And I have to say, now I used to get drafts for papers. I'd be pulling the hair out of my head, because I wouldn't even know where to start editing. But now I can focus on the underlying ideas rather than on the construction of the sentences, the coherence between one sentence and the next, and paragraph structure, because they run it, presumably. I haven't asked that. I should ask, you know, such as ChatGPT and at least, you know, it's decently written. Anyway, before we end, and I think we're running over time here, I'd like to briefly turn to the subject of scholarship. You've received grants from the National Endowment for the Humanities. I've received grants from the National Science Foundation and other federal agencies that have permitted us to engage in scholarship. So how do you think Genitive AI will impact research and scholarship? And maybe in two ways. are two sort of sides to this question. One is, will it lead to new forms of scholarship? Will we be able to use it to analyze things that are too big or complex for the human mind to analyze? That's one part of the question. And the other is, will this make it easier for people to access and understand knowledge and information produced by scholarship?

Jonathan Malesic (30:32)
Yeah, so I have I have a quick answer, but I want to turn the question around to you also. So my quick answer is that I can think of, know, so I'm, you know, I'm PhD is in religious studies. So I work with with texts, primarily. And, you know, I wrote a paper some years ago, where I was really aided, but I was trying to look up this one word in Danish in a book by the philosopher Søren Kierkegaard and see all the different ways he uses it. And I used, you know, I just did like a, you know, control F basically to find all of those and document them. That's a task generative AI would have been able to do in seconds. And You know, I was aided by, you know, the digitization of this book that I had already begun that work by just circling everything, you every instance of the word. and so I'm, I am certain that there are applications, you know, in, in the humanities, but I want to, the thing I want to turn around back to you is, you know, how does it affect the kind of, well, what's the term? you know, least publishable result or something like that in the sciences, you know, these teeny tiny correlations that can get turned into papers, then perhaps the paper can be generated basically overnight.

Eric Mazur (31:56)
Yeah. Well, I think, you know, I'm thinking about bigger things. mean, science is to some degree about pattern recognition and deep, deep, deep down, it's not that different from the arts or the humanities even. I think that there's really a connection between the two. But science, I mean... a lot of different, in my own field, physics are really about pattern recognition. And patterns often elude the human mind and are picked up more easily by generative AI. So I've been to a couple of talks where people actually use generative AI to produce new knowledge and to find things that have eluded people for centuries. So personally, I think, yeah.

Definitely, this might actually advance our fields too. In the same way that digitization of Kierkegaard's book helped you find the many occurrences of a particular word.

Anyway, interesting times we're living in. It's an exciting time. And I think that what you said at the beginning, that deep down, students are devoted to learning. And it is the extrinsic motivation that we impose on them, exams, grades, that sort of, you know, forces them to take shortcuts.

Jonathan Malesic (32:55)
Yeah, we are certainly.

Eric Mazur (33:22)
And if we can somehow reignite that intrinsic desire to learn that we're all born with, know, then things will be good. John, thank you so much for an absolutely amazing discussion. I think we could have gone on for another couple of hours. So I would like to conclude by thanking our audience for listening and invite everybody to return to our next episode. Thank you so much, John, for this.

Jonathan Malesic (33:36)
Yeah. Thank you for having me.

Eric Mazur (33:50)
You can find our social learning amplified podcast and more on perusall.com slash social learning amplified or one word subscribe to find out about other episodes. And I hope to welcome you back on a future episode.