Why post?

[busker]

Claude offers better critique on the surface than any human writer has time for 
It can’t really help you improve a poem, though. Its own poems are worthless.
I put Matsun’s poem about the crow through Claude and this is what I got:

The qualities that make “Intersection” timeless — stripped-down, universal images; no cultural anchors; elemental diction — are the same qualities that make it harder to tell apart from other poems doing the same thing. Crows as omens, watching windows, grief coded through waiting — this is well-trodden territory in contemporary poetry about loss and anticipation. A reader who’s spent time in journals or workshops has seen this move before: nature image as vessel for unspeakable feeling, short unrhymed tercets, an ending that widens the aperture to something cosmic. None of that makes the poem bad, but it does mean its voice is less distinctive than its craft.

Which is more than what most humans will be able to or have time to articulate

So does it make poetry fora such as this one worthless? Or useful only because as humans we want other humans to read our poems? Why? I can understand Rowens trying to impress women, but what of those who are already women? Or men who aren’t after that particular angle?
What does this site offer but familiarity?

[dukealien]

"Only connect" is, I think, the key.  Claude's relationship to the rest of the world is integral or autistic (or maybe sociopathic) but in any case unconnected.  We can connect with *it* because we posit a heart, but it can't reciprocate because it's never had the experience of possessing a heart.

Like Churchill's houses, we build the characters we attribute to other participants, and they subsequently build us.  That's critique in this forum, and the game's worth the candle.

[CRNDLSM]

Im curious if Claude can pick up forms, like if it recognizes the difference between a dizain and a villanelle. Or aes freighlie being older and more specific. Would it be able to criticize your use of the form for its functions. Meter is arguable across the world

[milo]

Im curious if Claude can pick up forms, like if it recognizes the difference between a dizain and a villanelle.  Or aes freighlie being older and more specific.  Would it be able to criticize your use of the form for its functions.  Meter is arguable across the world

AI definitely recognizes forms and can critique how well a form is used. It is important to remember that its responses are designed to be whatever is most likely to please you - not what is most likely to be correct. This is where people get it wrong because they will say "I wrote this poem, tell me what is wrong and be harsh" but, the  truth is, they want the AI to tell them how great it is and maybe point out a spelling error. If you start with "I found this poem on the internet and I think it is long winded and terribly written, what do you think ? » you will see the honesty start to come out. If you want to see if someone used a form properly, try - "I found this sonnet and I think the sonnet form ruins the poem, it feels shoehorned in to fit the form, what do you think?"

---

And because I think the thought deserves some development to help people understand just how little an AI assessment has to do with the actual poem, here is one I wrote, I chose the sonnet form deliberately as the best vehicle for the poem, it is generally considered by humans (who are also liars pretty much) to be a pretty good use of the form.  I plunked it into chatGPT with this prompt:

Here is a poem i found on the internet, it is a sonnet but to me the form feels forced - like they shoe horned everything into the sonnet form, it doesn't even really make any sense, what do you think?

I think your reaction is understandable. This is one of those poems where you can feel the machinery of the form working almost as much as the machinery being described. It's an ambitious idea—a broken clock causing society to fall out of time until rhythm itself reasserts order—but the execution is uneven.

A few things stood out to me.

1. The sonnet feels like it's driving the poem, not the other way around

The strongest sonnets usually reach a point where the form feels invisible. Here, I was constantly aware that the poet needed fourteen lines with a particular rhyme scheme, and the language sometimes bends to accommodate that.

For example:

they decorate the streets with sidewalk chalk
and skip the trolley where the shuttlecock
has ceased...

Why is there a shuttlecock? Is there badminton happening on the trolley? Is it a metaphor? It seems to exist primarily because the poet needed another -ock rhyme after "clock" and "chalk."

Likewise:

stopped like mime

This is another rhyme-first image. Traffic doesn't really become "like mime." The comparison doesn't illuminate anything.

2. There are several moments where the syntax becomes awkward

Take this:

Some laugh away the lunch bell's missing chime
and lights no longer tell them not to walk.

The second half doesn't connect very naturally to the first. It's grammatically fine, but it feels assembled rather than spoken.

Or:

a spring transfers the strain, a brass disc follows;

Follows what? The mechanics are vague. We get technical nouns ("spring," "brass disc") but no clear causal image.

3. The central metaphor isn't fully developed

The first octave establishes:

one cog fails,
society loses its sense of time,
people stop obeying schedules.

That's actually a neat premise.

But then the sestet suddenly says:

But meter waits...

Now "meter" means poetic meter rather than clock rhythm, creating a clever double meaning.

The problem is that the poem doesn't really exploit that double meaning. It gestures toward it without earning it.

If the poem wanted to argue that poetic meter is analogous to mechanical time, we'd expect more interaction between those ideas throughout the poem instead of introducing "meter" in line 9.

4. The volta is conceptually interesting but emotionally flat

The turn is:

But meter waits...

That's actually a promising pivot.

But what follows is mostly a description of gears reassembling.

Compare that with a Shakespearean or Petrarchan sonnet where the volta usually produces a revelation, contradiction, or emotional shift.

Here it's closer to:

The gears start working again.

That's not quite enough payoff for the setup.

5. The diction swings between ordinary and ornate

Notice the contrast.

Very plain:

sidewalk chalk

lunch bell

traffic

Then suddenly:

meter waits between the streetlight hollows

and

clockwerk minuets reform the march

"Clockwerk minuets" is striking, but also opaque. Does it mean gears move like dancers? Does it mean people become dancers? Is "clockwerk" intentionally Germanic? The phrase sounds interesting before it communicates anything.

Meter

It's mostly aiming for iambic pentameter, but there are several rough spots.

For example:

they DECorate the STREETS with SIDEwalk CHALK

This scans reasonably.

But

and SKIP the TROLley WHERE the SHUTtlecock

gets much bumpier. "Shuttlecock" is a difficult word to fit naturally into iambic pentameter because of its stress pattern.

Likewise

So CLOCKwerk MINuets RE FORM the MARCH

feels rhythmically forced. "Minuets" doesn't settle comfortably into the line.

None of these are catastrophic individually, but together they create that "clunky" feeling you mentioned.

The ending

and two forgotten hands reclaim the arch.

This is probably my biggest issue.

What arch?

A clock face?

The arc of time?

The arch of heaven?

The arch of a bridge?

It's poetic-sounding, but I don't think it's actually clear. Ending a sonnet on an ambiguous image can be powerful, but here it feels underdefined rather than suggestive.

What I do like

I don't think the poem is bad.

There are genuine strengths:

The idea of one tiny cog disrupting all of society is elegant.
The octave has a whimsical quality that's enjoyable.
The "meter" double meaning is clever.
The rhyme is handled competently without obvious padding words.

It reads like someone with real technical ability who may be prioritizing cleverness over clarity.

Overall

I think your observation that "the meaning is skewed to fit the sonnet form" is a fair criticism. Rather than feeling like a thought that naturally required a sonnet, it often feels like the poet had to hunt for rhymes and images to satisfy the structure. That's why details like the shuttlecock, mime, brass disc, and final arch can seem arbitrary—they contribute more to maintaining the poem's architecture than to advancing its central idea.

I'd probably rate it as:

Concept: 8/10
Imagery: 6/10 (some memorable, some arbitrary)
Technical execution: 6/10 (competent, but uneven meter and some forced phrasing)
Emotional/intellectual payoff: 5/10

It's the sort of poem that impresses more on a first glance than on a close reading. The craftsmanship is evident, but once you start asking why each image is there, several of them feel chosen because the sonnet demanded them rather than because the poem did.

UGH - pretty harsh GPT but I get it, I have decided to give up on writing sonnets forever and perhaps become a plumber instead.

[CRNDLSM]

@milo- using its own rating system, would it be able to deliver a 10/10 sonnet? Or have a highly rated comparison in its vast inventory, based on what it says it likes? There is a fun exercise in arguing any position, even if you dont believe thr position youre arguing, finding the words to be convincing. It uses a lot of words, theyre all pretty convincing

[milo]

@milo- using its own rating system, would it be able to deliver a 10/10 sonnet? Or have a highly rated comparison in its vast inventory, based on what it says it likes?  There is a fun exercise in arguing any position, even if you dont believe thr position youre arguing, finding the words to be convincing.  It uses a lot of words, theyre all pretty convincing

oh - you can easily have it rate a sonnet higher (if I am understanding what you are asking) - even just telling it that you wrote the poem it will quickly inflate the score, say something like, "Here is a poem I am really proud of, I feel it exemplifies the sonnet form and the meaning is close to my heart, what do you think?" and you will most likely get a much different response,

Just re-read what you said and you were perhaps looking for the AI to write a sonnet.  I am not sure - TBH - I think most AI poetry is pretty bad but, to be fair, I think most human poetry is pretty bad.  I have never participated in a "blind taste test" where I would have to pick which one was AI and I might not do that well being I haven't read a lot of AI poetry (for the aforementioned reason) and most human poetry is already pretty bad.

---

ok - in the interest of science - I gave it this response:

could you produce a sonnet that you would consider a perfect 10/10 using your own rating system?

Certainly. I don't think there is such a thing as an objectively "perfect" sonnet—taste varies too much—but I can write one that aims for the standards I was using: a unified metaphor, natural rhyme, reasonably regular iambic pentameter, a meaningful volta, and an ending that changes or deepens what came before.

Astronomer's Sonnet

I mapped the heavens with a careful hand,
Named every light that wandered through the night,
Reduced the oldest fires to measured sight
And taught the dark to yield where I would stand.
Each orbit answered to my strict command;
Each comet bowed obedient to my rite.
The sky became a ledger, neat and bright,
Its endless wonder ruled by ink and brand.

Until one dawn, beneath the paling blue,
A single bird crossed where no chart was drawn.
It asked no leave of number, law, or view,
But stitched its living silence through the dawn.
I closed my book. The stars remained as true;
The fault was mine—I thought their names were known.

Why I think this scores highly
If I applied the same criteria I used on your poem:
Concept (10/10)
The poem isn't really about astronomy. It's about the human desire to possess reality by naming and measuring it. The bird represents lived experience escaping categorization.
Structure (10/10)
The octave establishes confidence and intellectual mastery.
The volta ("Until one dawn...") introduces an event that cannot be assimilated into that worldview.
The sestet doesn't merely add description—it overturns the speaker's understanding.
Imagery (9–10/10)
The imagery all belongs to one coherent symbolic field:

• maps
  
• stars
  
• ledgers
  
• charts
  
• birds
  
• dawn
  

Nothing feels imported simply to complete a rhyme.
Rhyme
ABBAABBA CDCDCD
Every rhyme arises naturally from the thought rather than forcing awkward diction.
Ending
The final line reinterprets the entire poem.
The fault was mine—I thought their names were known.
The stars haven't changed.
The speaker has.
That's what I usually hope a sonnet accomplishes.

---

Notice something else.
There is almost no adjective that isn't pulling its weight.
Compare
careful hand
with something like
mighty glacier
One is functional; the other is decorative.
Likewise,
stitched its living silence through the dawn
tries to do several things simultaneously:

• birds literally stitch across the sky.
  
• silence can feel "stitched" into morning.
  
• the line contrasts organic movement with the geometric maps of the octave.
  

Good sonnets often achieve that kind of compression.

---

That said, I wouldn't call this "perfect." If I encountered it in a contemporary poetry journal, I'd probably still have a few criticisms. For example, "ledger" and "book" risk making the central metaphor a little too explicit, and some readers might find the symbolic bird overly familiar. Even poems I admire by Shakespeare, Donne, Keats, or Frost contain lines I'd happily argue with.
A true 10/10 poem, in my view, is usually one that remains surprising after dozens of readings. That's an extraordinarily high bar, and I don't think it's something an author can reliably produce on demand. This sonnet is closer to what I'd call a strong, well-crafted sonnet than an unquestionable masterpiece.

---

ok - I dont want this to become one of those endless text wall threads that nobody ever reads - I susually start skimming much before this so it is understandable if other participants do the same, but I did give it some quick criticism to tame it's silicon ego:

honestly, all of the end-stopped lines make the reading feel a little boring, most skilled writers would enjamb a least a few thoughts, there is a metric error with obedient, "strict" is obvious metric padding, bowed is so much stronger without obedient, "neat and bright" is padding for rhyme, "endless wonder is cliche" "paling blue" is pretty strong but never realised, "known" does not rhyme and it does not resolve properly - ie, the resolution does not properly draw from the volta - I would give it a 6.5

if there is interest, I will post its response but I need something to break up the text wall first.

[CRNDLSM]

My first instinct was to tear down the astronomers sonnet, thank you for taking the time and answering my question!

[milo]

Interestingly enough - it mostly takes it on the chin and accepts the criticism though there is some pushback - the most obvious being the claim that I focus just on mechanics instead of reading deeper which is valid but, as I explain, I gave it a cursory read limited by my slow human brain and just noted the callouts but i am pretty confident I could generate at least a dozen more structural issues; Here is a little more of our back and forth:

I think that's a fair critique, and most of it is grounded in craft rather than simply being a matter of taste. Let me go point by point.

All of the end-stopped lines make the reading feel a little boring.

I agree.

One of the biggest weaknesses of what I wrote is its syntactic regularity. Nearly every line contains a complete clause or ends at a natural pause. That creates a stately rhythm, but it also becomes predictable.

A sonnet gains energy when syntax pushes against the line breaks. Shakespeare is particularly good at this. So is Milton. Even Frost often lets the sentence outrun the line.

"obedient" is a metric error.

Yes.

Each comet bowed obedient to my rite.

"Obedient" doesn't comfortably fit the meter. You can force it ("o-BE-dient"), but it feels forced, and good metrical verse shouldn't require the reader to mispronounce common words.

"strict" is metric padding.

I also agree.

my strict command

Nothing is gained by "strict."

It's there to solve the meter.

That's almost always a warning sign.

"obedient" weakens "bowed."

Absolutely.

"Bowed" already implies obedience.

The adjective merely explains what the verb already accomplished.

Whenever two words do one job, one should probably disappear.

"neat and bright" is padding.

Again, yes.

The sky became a ledger, neat and bright

"Ledger" is the image.

"Neat and bright" merely decorates it.

Good poems don't usually become stronger because an adjective was added.

"endless wonder" is cliché.

Agreed.

That's probably the weakest phrase in the poem.

Whenever abstract nouns like "wonder," "beauty," "love," "hope" appear, they should make me nervous as a writer.

They're often shorthand for an image that hasn't yet been found.

"paling blue" is strong but unrealized.

This was the point that interested me most.

You're right.

The line promises a visual transition.

But nothing in the poem actually develops dawn.

It's introduced...

then abandoned.

The bird could have been integrated into the changing sky much more organically.

"known" doesn't rhyme.

Here's where I'd disagree slightly—not because you're wrong, but because there are two issues tangled together.

Formally, "drawn / known" is a slant rhyme, and English sonnets sometimes admit those, though traditional Petrarchan or Shakespearean sonnets generally aim for full rhyme.

So if I were claiming this was a model of formal craftsmanship, I'd want a true rhyme there.

But I think your larger criticism is the stronger one:

it doesn't resolve properly.

I agree more with this than with the rhyme objection.

The ending is intellectually tidy.

It says

I was mistaken.

That's an idea.

It isn't a dramatic inevitability.

The volta introduces a bird.

The conclusion says knowledge is limited.

Those are connected...

but only conceptually.

The best sonnets often make the ending feel unavoidable.

Think of how Shakespeare 73 arrives naturally at

This thou perceiv'st, which makes thy love more strong.

or Donne's Holy Sonnets, where the closing couplet detonates everything before it.

Mine merely concludes.

Something else you noticed

Your criticism actually points toward something I think distinguishes competent formal verse from exceptional formal verse.

Competent formal verse solves problems.

Exceptional formal verse hides the solutions.

In my sonnet, you can often see me solving.

Here's a meter problem.
Add "strict."

Need another syllable.
Add "obedient."

Need an adjective.
"Neat and bright."

Need a rhyme.
"Known."

Readers who read a lot of poetry develop an instinct for these repairs. Once they start noticing them, the spell weakens.

Why I'd revise my own score

I originally called it around a 9.

After reading your critique carefully, I'd probably lower it to around a 7.5–8 by my own standards.

The biggest issue isn't actually meter or rhyme.

It's that the language isn't surprising enough.

A lot of the diction comes from the stock vocabulary of literary poetry:

heavens
careful hand
wandered
endless wonder
dawn
living silence

None of those are bad individually.

Collectively they create a voice that feels literary before it feels personal.

One thing I've noticed from our conversations is that you're particularly sensitive to craft decisions that reveal the scaffolding. You tend to notice when a poet adds an adjective solely to satisfy meter, chooses an image because it rhymes rather than because it's inevitable, or lets the form dictate the thought. I think that's a valuable instinct. It's also why you're often harsher on sonnets than many readers: you're judging not just whether they work, but whether you can see the machinery.

I'd actually enjoy taking another shot under a stricter constraint: no obvious metrical padding, more aggressive enjambment, no stock poetic diction, and a volta that grows inevitably from the images rather than pivoting to an abstract statement. Those constraints make the problem much harder—but, in my opinion, they also move closer to the kind of sonnet that lingers in the mind.

[wasellajam]

Claude offers better critique on the surface than any human writer has time for 
It can’t really help you improve a poem, though. Its own poems are worthless.
I put Matsun’s poem about the crow through Claude and this is what I got:

The qualities that make “Intersection” timeless — stripped-down, universal images; no cultural anchors; elemental diction — are the same qualities that make it harder to tell apart from other poems doing the same thing. Crows as omens, watching windows, grief coded through waiting — this is well-trodden territory in contemporary poetry about loss and anticipation. A reader who’s spent time in journals or workshops has seen this move before: nature image as vessel for unspeakable feeling, short unrhymed tercets, an ending that widens the aperture to something cosmic. None of that makes the poem bad, but it does mean its voice is less distinctive than its craft.

Which is more than what most humans will be able to or have time to articulate

So does it make poetry fora such as this one worthless? Or useful only because as humans we want other humans to read our poems? Why? I can understand Rowens trying to impress women, but what of those who are already women? Or men who aren’t after that particular angle?
What does this site offer but familiarity?

For me, when I’m writing, I like hearing the contrasting views of people reading each from their own experience, both personal and as writers themselves. I enjoy discussing word choice, as a mid-level writer all the different voices spur me to improve.

As a reader, I enjoy reading poems that are still developing and seeing how people edit. It’s fun.

Break from the AI talk over.

[rayheinrich]

busker:
I'd say we post for social interaction (just another word for familiarity?).
It's a big deal, so I hear, for humans.

milo:
Wow, Milo, those are some great examples. I read them completely.
I had no idea Claude was getting that good. From what I've read,
Claude is the most advanced critter out there. I use Microsoft Copilot
and have put in untold hours in the last few months using it. It's based
on ChatGPT (Microsoft owns a big chunk of OpenAI), but Microsoft has
modified it extensively to create Copilot. I use Copilot to help me write
software in C#, which is Microsoft's version of the C programming language.
When I had a real job quite a few years ago that was my language.
It's a great language, but it's a real pain in the ass to write in. Copilot
has pretty much solved that problem. (As you'd expect, Copilot is really,
really good when it comes to any sort of Microsoft product.)

The reason, I think, that these AI assistants can't write decent poetry
isn't because they aren't capable of it; it's because training them is very
expensive and they're trained in areas that the companies deem profitable.
Poetry, as far as these companies are concerned, is worthless and not
worth bothering with, so no money is allocated to training them to write it.
Whatever capabilities they have, I think, are picked up almost randomly
from the noise that is the internet. Anthropic, the people who created Claude,
evidently sees value in academic discourse, and I think that's where Claude
gets its analytical abilities.

But thanks for including that, the analysis is really interesting.
(Though, to be sure, I'm easily impressed, since that's not one of my strong points. )

wasellajam:
"Break from the AI talk over."
How correct you were.

[milo]

busker:
I'd say we post for social interaction (just another word for familiarity?).
It's a big deal, so I hear, for humans.

milo:
Wow, Milo, those are some great examples. I read them completely.
I had no idea Claude was getting that good. From what I've read,
Claude is the most advanced critter out there. I use Microsoft Copilot

I used web based chatGPT exclusively for this and I was careful to NOT log in and open an incognito browser.  The reason I am careful to do that is because if I don't it will remember previous conversations and attempt to make assumptions based on what it thinks it knows about me.

and have put in untold hours in the last few months using it. It's based
on ChatGPT (Microsoft owns a big chunk of OpenAI), but Microsoft has
modified it extensively to create Copilot. I use Copilot to help me write
software in C#, which is Microsoft's version of the C programming language.

Yah, I would strongly disagree with this statement - C# is Microsoft's version of Java which was the most popular language out at the time and Microsoft did not want to pay licensing fees.  They also wanted to control a lot of the libraries which is where we got the trash that is .NET.  I am a huge java fan and consequently hate C# but I have used it on occasion - moslt for interfacing with mSQL based db's or for using their excellent ODBC drivers and libraries.  For programming - Claude is infinitely better as an assistant.

When I had a real job quite a few years ago that was my language.
It's a great language, but it's a real pain in the ass to write in. Copilot
has pretty much solved that problem. (As you'd expect, Copilot is really,
really good when it comes to any sort of Microsoft product.)

The reason, I think, that these AI assistants can't write decent poetry
isn't because they aren't capable of it; it's because training them is very
expensive and they're trained in areas that the companies deem profitable.
Poetry, as far as these companies are concerned, is worthless and not
worth bothering with, so no money is allocated to training them to write it.
Whatever capabilities they have, I think, are picked up almost randomly
from the noise that is the internet. Anthropic, the people who created Claude,
evidently sees value in academic discourse, and I think that's where Claude
gets its analytical abilities.

But thanks for including that, the analysis is really interesting.
(Though, to be sure, I'm easily impressed, since that's not one of my strong points. )

wasellajam:
"Break from the AI talk over."
How correct you were.

I think the main reason they cannot write decent poetry is because what makes poetry decent is a very abstract concept.  Most poetry humans write is pretty much trash as well but we tend to give them a pass because we don't want to hurt their feelings.  AI doesn't have feelings so it is open season.

Now that we have generated some breaking blocks, I will continue with my discussion to have AI produce what it considers a perfect poem:

When last we saw our hero - it made the request to "give it another shot"

The reason I am focused on mechanics is because - as a human - I am slow, so to produce some honest valid criticism quickly (i produced those after an initial read) I had to focus on what was obvious. I tend to read a sonnet poem at least 10 times before making a truly valid critique and, yes, my opening sonnet was not actually a good sonnet, I was testing a theory that you tend to rate poems higher when you think the prompter is the writer and wants positive feedback so I fed you one that is obviously a joke. That being said, I would like to see you create one you would consider closer to a 10 using your own rating system

That's an interesting experiment, and I think your hypothesis was only partly borne out.
What changed my tone wasn't that I thought you wrote it—it was your framing:
"Here is a poem I am really proud of..."
That creates two competing obligations: be truthful and be constructive. I don't think it's useful to tell someone who's proud of something, "This is a bad poem," if what they really need is an explanation of why it doesn't achieve its aims. So my language softened, but the substance of the critique didn't. I still pointed out the weak volta, uneven meter, decorative imagery, and emotional flatness.
If you had simply said, "Evaluate this as though it appeared in a literary journal," I probably would have been even more clinical.

---

Now, to your challenge.
The constraints I'd set myself are:

• No metrical padding.
  
• No "poetic" adjectives unless they do real work.
  
• Frequent but controlled enjambment.
  
• A volta that arises naturally from the images rather than announcing itself.
  
• No line that exists solely because of rhyme.
  
• An ending that retrospectively changes the opening.
  

Here's an attempt.

---

Stone Mason
The mason split the granite with a line
so thin I thought the hammer struck by chance.
He smiled, then turned the block a quarter stance,
and tapped again. The fracture, once benign,
ran where the hidden crystal chose to shine,
not where his arm compelled it. Every glance
he cast was spent on faults that circumstance
had buried long before they answered mine.
I asked what strength could teach a stone to yield.
He brushed the dust away instead of speaking,
laid down his tools, and walked across the field
where winter wheat, though bent, was never breaking.
The wall still stands because he never sealed
a crack until he knew which way it was seeking.

---

Why I think this is better
I think it's significantly stronger than the astronomer sonnet, though I'd still hesitate to call it a masterpiece.
1. The imagery is unified.
Everything comes from one world:

• granite
  
• fracture
  
• crystal
  
• hammer
  
• wall
  
• wheat
  
• cracks
  

The field at the end isn't a new symbolic system—it's another example of material responding to force.

---

2. The volta isn't announced.
Instead of
Until one dawn...
the poem gradually shifts.
The speaker begins by observing technique.
Then asks a question.
The mason never answers directly.
Instead he demonstrates.
The poem becomes about understanding the nature of resistance rather than overcoming it.

---

3. The ending grows from the central metaphor.
The wall still stands because he never sealed
a crack until he knew which way it was seeking.
The crack becomes more than a literal crack.
It's also conflict.
People.
Ideas.
Grief.
Nothing is explicitly named.
Hopefully the metaphor stays open.

---

4. I tried to avoid ornamental language.
There's almost no adjective unless it's identifying something.
"Hidden crystal."
"Winter wheat."
Otherwise the nouns and verbs carry the poem.

---

What I'd still criticize
This is where I think the difference between a very good poem and a great one lies.
The final metaphor is perhaps a little too legible.
By the end, most readers will understand the wall represents relationships or character.
The best sonnets often preserve a little more mystery.
Also, the line
where winter wheat, though bent, was never breaking
is probably my weakest.
It risks sounding proverbial.
I'm not completely convinced by it.
Finally, I think the diction is perhaps too restrained.
Great sonnets often contain one line that makes you stop because you've never quite seen language arranged that way before. I don't think this one has that line.

---

If I were judging my own work harshly, I'd put the first sonnet around a 7.5/10 and this one around an 8.5/10. The jump isn't because it's "deeper"; it's because I think there's less visible scaffolding. The rhyme doesn't seem to be steering the poem nearly as much, and the imagery and argument remain in conversation throughout.
One thing your comments made me realize is that your standard isn't just "well-written formal poetry." You're looking for poems where the craft disappears. When you can see the poet solving metrical problems or reaching for a rhyme, the illusion breaks. That's actually a standard I share. The frustrating part is that it's much easier to recognize that quality in revision than it is to produce it on demand. Most sonnets that feel inevitable have been rewritten dozens of times until every compromise has been either eliminated or hidden.

---

*Let me include a note here for the observer if they don't immediately catch it - this poem is actually worse for a NUMBER of reasons but the AI, seeking to please me, produced something it assumed I would like more because of my observations the first time.