It's still really completely useless at anything even slightly less known, and in annoying ways that take a minute to figure out (that it doesn't know anything and is genuinely confabulating); or at less standard takes. So for example, it deals with the pentancle symbolism and role in the "Gawain & Green Knight" ok, but it mostly falls on its digital face on the subject of the Green Chapel. And the reason for it seems pretty simple: much more had been written on the former than the latter.
Similar with close reading of more contemporary poetry. I'm not sure what would happen if it's given a text verbatim and some context/approach angle but I'm not holding my breath for eg metre analysis considering it's still incapable of consistently formatting citations in less frequently used styles.
I gave it Meadowsweet by Kathleen Jamie. Well known but not internationally revered poet, not one of her frequently analysed or iconic text; needs contextual knowledge for full understanding.
It worked from text alone and it provided a plausible general interpretation marred by a complete lack of contextual mooring. Nudged towards that in a follow up probing, it actually did improve.
Better on meanings than techniques spontaneously, but when explicitly requested techniques analysis, it's actually done VERY well (for a robot which *knows* nothing).
All in all, I'm positively surprised especially with the results of the techniques prompt. With guidance / right prompting it could potentially provide genuinely useful output. Obviously one needs to know what questions to ask and how to get past the generic "human conditions then we die" stuff, but something usable definitely emerging here.
It still doesn't know "honesty" is a plant, but that's forgivable ;)
Please respect writers. When you "give" someone else's writing to AI, you are giving away intellectual and artistic property you do not have rights to. Did that writer intend their work to be chopped into bits and spit out in a mashup with other scraped garbage?
The poem was freely available in public domain without any paywalls, so while pointing to it made it more likely that the LLM would process it, it's not a text that wasn't "out there" already. But in the future I will stick to long-dead authors for such purposes.
A challenge today is that authors who agreed to make work open access in order for people without access to expensive databases did NOT anticipate AI or have in mind that bilionaires would take their work. Creative Commons is wrestling with this question and trying to come up with new licenses that give writers more options. That said, given that AI doesn't respect copyright, would they respect CC? In the meantime, respect writers and don't upload their work.
If it's getting "better," that's probably because some academic publishers are signing deals allowing AI companies to train their software on their authors' work. About twice a week I get an email from Cambridge UP reminding me to sign the updated contract giving away my rights in this respect. They don't offer anything for this, nor do they offer the chance to refuse. I've written their staff and told them I will not sign this, they've apologized and promised the emails will stop, the emails did not stop. So lovely that my publisher is keen on selling away rights to my work and increasing the speed according to which my students stop reading and writing.
In short: AI did not produce that "knowledge." People did. But people won't anymore, not the next generations, because they will be so reliant on this software that they will neither be able to read full texts for themselves (this is already a challenge) nor will they be able to write a few hundred words of their own to figure out their own thoughts. I hope someone is having fun with this.
"All that vague and unknowledgeable talk about feelings and impressions..." But that's the entry point, isn't it? Feelings and impressions. That's how we first engage with literature, with poetry. It's how most mere mortals enjoy reading. Machines can't enjoy it, can they? They can amass knowledge until the end of time but they're never going to be able to enjoy it. We respond to poetry, before we can name any of the ways it achieves its effects. It's alive to us. There's an aridity to literary criticism done by a machine. It's only ever going to be able to pin the butterfly to the wall and label it. And that material will be good enough for many students of literature in the future. It's not going to be 'improving opinion into knowledge' any more that the Chorley Method improves bread (mixing my metaphors from butterflies to bread here). AI is just going to make a load of product, that's all. Literary criticism UPF. I just don't see the point in talking up the things AI can do now in literary criticism when it can never ever experience human feeling which is the starting point of literature, and the animating drive behind all Art.
oh of course, and I think I have written about that before, but on the narrow problem of "AI as literary critic" if it is going to be better/good it needs to get away from that sort of talk. Who needs AI to waffle on about feelings? Even if all it ever does is what you describe, we want it to be good at that. It's here and it's not going away!
"What on earth does it mean to be fully human? No-one can say. " I can. You breathe and have a beating heart. I, a human, will read books written by humans, and my mind and heart will be better for it.
Sorry Henry, can't get excited about AI criticizing literature it pirated or stole. Using untold environmental resources (cue pictures of LA and look at water uses of AI) and doing what exactly to make the world a better place?
Wow. It's really interesting to see the comments here. It probably shows how this thread is removed from my normal substack reading material because I've genuinely never before seen this level of hostility (defensiveness?) towards AI tools not driven by fear of a lethal singularity.
Wait. Actually I have. It's comparable to what I saw start on translators' fora a few years ago when MT started to really take off with noticeably improving quality.
i use it most as, as you mention above, a quick explainer. also, there's an extension you can use called "sider" that will 'explain' or 'translate' anything you highlight on a page. right now, finding it pretty handy (on the free tier). seems to read around what you highlight to get any context. reasonably smart and you can change the model it uses.
The first answer from Google Gemini - 2.0 Experimental Advanced. Lacks access to real-time info and some Gemini features.
The statement "Whatever else its merits, The Clerk's Tale does not achieve its effects through surprise" is a compelling, albeit potentially debatable, assertion. While the narrative of "The Clerk's Tale" from Chaucer's Canterbury Tales possesses numerous merits, including its moral complexity and philosophical depth, it's largely true that surprise isn't its primary tool for engaging the audience. This is primarily due to the tale's reliance on pre-existing narrative models and its focus on didacticism over dramatic twists. However, a nuanced discussion reveals that while dramatic surprises may be absent, the tale does employ subtle forms of surprise on an emotional and intellectual level.
Firstly, the tale's lack of reliance on surprise is evident in its source material. The Clerk explicitly states that he learned the story from Petrarch, who adapted it from Boccaccio's Decameron. The story of patient Griselda was a popular medieval tale, a fact acknowledged by the Clerk himself. Therefore, many in Chaucer's audience would have been familiar with the basic plot: a nobleman tests his wife's obedience through increasingly cruel trials. Knowing the broad strokes of the narrative beforehand diminishes the potential for the kind of shocking twists that characterize tales like the Miller's or the Reeve's. The audience anticipates Griselda's suffering and Walter's eventual revelation, rendering the plot points themselves less surprising.
Furthermore, the Clerk's didactic purpose overshadows any desire to create suspense. He presents the tale as an exemplum, a story meant to illustrate a moral lesson. The focus is on Griselda's unwavering obedience and patience, which are presented as ideal, albeit extreme, virtues. The Clerk even interrupts the narrative with moralizing asides, guiding the reader's interpretation. This emphasis on didacticism means that the narrative is structured to reinforce a pre-determined message rather than to surprise the audience with unexpected developments. The trials inflicted upon Griselda, while horrifying, serve to showcase her steadfastness, a quality the Clerk openly admires. The predictability of the narrative structure, therefore, contributes to the lack of dramatic surprise.
However, to say that the tale achieves no effects through surprise is an oversimplification. While plot twists may be absent, the tale employs psychological and emotional surprise. The sheer extremity of Walter's cruelty often shocks the audience, even if they anticipate the general trajectory of the plot. The removal of Griselda's children and her public humiliation are designed to elicit strong emotional responses, bordering on outrage and disbelief. This cruelty pushes the boundaries of acceptability, creating a sense of unease and questioning the motives of the character, even if his actions are somewhat foreshadowed. Thus, the surprise lies not in what happens, but in how far Walter takes his tests, constantly escalating the emotional stakes.
Moreover, the tale generates a kind of intellectual surprise through its ambiguity. The Clerk's interpretation of the tale as an allegory for Christian faith is not entirely convincing, and the ending, where he distances himself from the extreme implications of Griselda's behavior, leaves the audience with unresolved questions. Is Griselda a praiseworthy example of virtue, or a victim of senseless abuse? Is Walter a tyrannical figure, or a flawed man testing the limits of human endurance? These open questions force the audience to engage with the tale's moral complexities, creating a different kind of surprise – the surprise of unexpected moral ambiguity in a seemingly straightforward story. The Clerk's own ironic tone, especially in his envoy, further complicates the intended message, leaving space for multiple interpretations and contributing to this intellectual surprise.
In conclusion, "The Clerk's Tale" primarily achieves its effects through means other than dramatic surprise. Its reliance on a familiar narrative, its didactic focus, and its predictable structure all diminish the possibility of surprising the audience with plot twists. However, the tale does generate surprise through the extremity of Walter's cruelty, the unexpected emotional impact of Griselda's suffering, and the lingering intellectual ambiguity surrounding its moral message. While not a tale of suspense in the conventional sense, "The Clerk's Tale" still manages to surprise its audience, albeit in a more subtle and thought-provoking manner. It forces us to confront uncomfortable questions about obedience, faith, and the nature of human relationships, a kind of surprise that lingers long after the story's end.
It's still really completely useless at anything even slightly less known, and in annoying ways that take a minute to figure out (that it doesn't know anything and is genuinely confabulating); or at less standard takes. So for example, it deals with the pentancle symbolism and role in the "Gawain & Green Knight" ok, but it mostly falls on its digital face on the subject of the Green Chapel. And the reason for it seems pretty simple: much more had been written on the former than the latter.
Similar with close reading of more contemporary poetry. I'm not sure what would happen if it's given a text verbatim and some context/approach angle but I'm not holding my breath for eg metre analysis considering it's still incapable of consistently formatting citations in less frequently used styles.
oh sure, I don't think it's going to be a great critic any time soon, but... something significant has changed
Actually I'm going to test it rn.
tell me what you find!
I gave it Meadowsweet by Kathleen Jamie. Well known but not internationally revered poet, not one of her frequently analysed or iconic text; needs contextual knowledge for full understanding.
It worked from text alone and it provided a plausible general interpretation marred by a complete lack of contextual mooring. Nudged towards that in a follow up probing, it actually did improve.
Better on meanings than techniques spontaneously, but when explicitly requested techniques analysis, it's actually done VERY well (for a robot which *knows* nothing).
All in all, I'm positively surprised especially with the results of the techniques prompt. With guidance / right prompting it could potentially provide genuinely useful output. Obviously one needs to know what questions to ask and how to get past the generic "human conditions then we die" stuff, but something usable definitely emerging here.
It still doesn't know "honesty" is a plant, but that's forgivable ;)
Please respect writers. When you "give" someone else's writing to AI, you are giving away intellectual and artistic property you do not have rights to. Did that writer intend their work to be chopped into bits and spit out in a mashup with other scraped garbage?
Good point - this is why it is important to have data sharing turned off in chatGPT. I encourage everyone to do that.
That's a valid point, thank you for making it.
The poem was freely available in public domain without any paywalls, so while pointing to it made it more likely that the LLM would process it, it's not a text that wasn't "out there" already. But in the future I will stick to long-dead authors for such purposes.
A challenge today is that authors who agreed to make work open access in order for people without access to expensive databases did NOT anticipate AI or have in mind that bilionaires would take their work. Creative Commons is wrestling with this question and trying to come up with new licenses that give writers more options. That said, given that AI doesn't respect copyright, would they respect CC? In the meantime, respect writers and don't upload their work.
If it's getting "better," that's probably because some academic publishers are signing deals allowing AI companies to train their software on their authors' work. About twice a week I get an email from Cambridge UP reminding me to sign the updated contract giving away my rights in this respect. They don't offer anything for this, nor do they offer the chance to refuse. I've written their staff and told them I will not sign this, they've apologized and promised the emails will stop, the emails did not stop. So lovely that my publisher is keen on selling away rights to my work and increasing the speed according to which my students stop reading and writing.
In short: AI did not produce that "knowledge." People did. But people won't anymore, not the next generations, because they will be so reliant on this software that they will neither be able to read full texts for themselves (this is already a challenge) nor will they be able to write a few hundred words of their own to figure out their own thoughts. I hope someone is having fun with this.
"All that vague and unknowledgeable talk about feelings and impressions..." But that's the entry point, isn't it? Feelings and impressions. That's how we first engage with literature, with poetry. It's how most mere mortals enjoy reading. Machines can't enjoy it, can they? They can amass knowledge until the end of time but they're never going to be able to enjoy it. We respond to poetry, before we can name any of the ways it achieves its effects. It's alive to us. There's an aridity to literary criticism done by a machine. It's only ever going to be able to pin the butterfly to the wall and label it. And that material will be good enough for many students of literature in the future. It's not going to be 'improving opinion into knowledge' any more that the Chorley Method improves bread (mixing my metaphors from butterflies to bread here). AI is just going to make a load of product, that's all. Literary criticism UPF. I just don't see the point in talking up the things AI can do now in literary criticism when it can never ever experience human feeling which is the starting point of literature, and the animating drive behind all Art.
oh of course, and I think I have written about that before, but on the narrow problem of "AI as literary critic" if it is going to be better/good it needs to get away from that sort of talk. Who needs AI to waffle on about feelings? Even if all it ever does is what you describe, we want it to be good at that. It's here and it's not going away!
"What on earth does it mean to be fully human? No-one can say. " I can. You breathe and have a beating heart. I, a human, will read books written by humans, and my mind and heart will be better for it.
Sorry Henry, can't get excited about AI criticizing literature it pirated or stole. Using untold environmental resources (cue pictures of LA and look at water uses of AI) and doing what exactly to make the world a better place?
Wow. It's really interesting to see the comments here. It probably shows how this thread is removed from my normal substack reading material because I've genuinely never before seen this level of hostility (defensiveness?) towards AI tools not driven by fear of a lethal singularity.
Wait. Actually I have. It's comparable to what I saw start on translators' fora a few years ago when MT started to really take off with noticeably improving quality.
i use it most as, as you mention above, a quick explainer. also, there's an extension you can use called "sider" that will 'explain' or 'translate' anything you highlight on a page. right now, finding it pretty handy (on the free tier). seems to read around what you highlight to get any context. reasonably smart and you can change the model it uses.
should add mostly i ask it pretty basic questions
"Talked a lot of flannel"! Never heard this and it's so apt and funny!
Old school
I'm going to use it when writer friend starts talking "theory."
I guess this is sightly reassuring...
The first answer from Google Gemini - 2.0 Experimental Advanced. Lacks access to real-time info and some Gemini features.
The statement "Whatever else its merits, The Clerk's Tale does not achieve its effects through surprise" is a compelling, albeit potentially debatable, assertion. While the narrative of "The Clerk's Tale" from Chaucer's Canterbury Tales possesses numerous merits, including its moral complexity and philosophical depth, it's largely true that surprise isn't its primary tool for engaging the audience. This is primarily due to the tale's reliance on pre-existing narrative models and its focus on didacticism over dramatic twists. However, a nuanced discussion reveals that while dramatic surprises may be absent, the tale does employ subtle forms of surprise on an emotional and intellectual level.
Firstly, the tale's lack of reliance on surprise is evident in its source material. The Clerk explicitly states that he learned the story from Petrarch, who adapted it from Boccaccio's Decameron. The story of patient Griselda was a popular medieval tale, a fact acknowledged by the Clerk himself. Therefore, many in Chaucer's audience would have been familiar with the basic plot: a nobleman tests his wife's obedience through increasingly cruel trials. Knowing the broad strokes of the narrative beforehand diminishes the potential for the kind of shocking twists that characterize tales like the Miller's or the Reeve's. The audience anticipates Griselda's suffering and Walter's eventual revelation, rendering the plot points themselves less surprising.
Furthermore, the Clerk's didactic purpose overshadows any desire to create suspense. He presents the tale as an exemplum, a story meant to illustrate a moral lesson. The focus is on Griselda's unwavering obedience and patience, which are presented as ideal, albeit extreme, virtues. The Clerk even interrupts the narrative with moralizing asides, guiding the reader's interpretation. This emphasis on didacticism means that the narrative is structured to reinforce a pre-determined message rather than to surprise the audience with unexpected developments. The trials inflicted upon Griselda, while horrifying, serve to showcase her steadfastness, a quality the Clerk openly admires. The predictability of the narrative structure, therefore, contributes to the lack of dramatic surprise.
However, to say that the tale achieves no effects through surprise is an oversimplification. While plot twists may be absent, the tale employs psychological and emotional surprise. The sheer extremity of Walter's cruelty often shocks the audience, even if they anticipate the general trajectory of the plot. The removal of Griselda's children and her public humiliation are designed to elicit strong emotional responses, bordering on outrage and disbelief. This cruelty pushes the boundaries of acceptability, creating a sense of unease and questioning the motives of the character, even if his actions are somewhat foreshadowed. Thus, the surprise lies not in what happens, but in how far Walter takes his tests, constantly escalating the emotional stakes.
Moreover, the tale generates a kind of intellectual surprise through its ambiguity. The Clerk's interpretation of the tale as an allegory for Christian faith is not entirely convincing, and the ending, where he distances himself from the extreme implications of Griselda's behavior, leaves the audience with unresolved questions. Is Griselda a praiseworthy example of virtue, or a victim of senseless abuse? Is Walter a tyrannical figure, or a flawed man testing the limits of human endurance? These open questions force the audience to engage with the tale's moral complexities, creating a different kind of surprise – the surprise of unexpected moral ambiguity in a seemingly straightforward story. The Clerk's own ironic tone, especially in his envoy, further complicates the intended message, leaving space for multiple interpretations and contributing to this intellectual surprise.
In conclusion, "The Clerk's Tale" primarily achieves its effects through means other than dramatic surprise. Its reliance on a familiar narrative, its didactic focus, and its predictable structure all diminish the possibility of surprising the audience with plot twists. However, the tale does generate surprise through the extremity of Walter's cruelty, the unexpected emotional impact of Griselda's suffering, and the lingering intellectual ambiguity surrounding its moral message. While not a tale of suspense in the conventional sense, "The Clerk's Tale" still manages to surprise its audience, albeit in a more subtle and thought-provoking manner. It forces us to confront uncomfortable questions about obedience, faith, and the nature of human relationships, a kind of surprise that lingers long after the story's end.