# ChatGPT Got Me Reading Plato
Matt Hodges
2025-09-23

Two weeks ago, the YouTube algorithmic gods suggested to me [The Gist Of
Western Philosophy](https://www.youtube.com/watch?v=7XcwlDnwai0). It’s a
great 15-minute video with the ambition of summarizing all of Western
philosophy. But what’s more impressive is what the creator claims
preceded the video: over the course of one year they read the works of
52 major philosophers.

<https://www.youtube.com/watch?v=7XcwlDnwai0>

Coming out of the video, I too wanted to become more acquainted with
philosophy. I read a little bit of **Kant** and **Bentham** (and I think
some **Foucault**?) in my undergraduate [Tech
Ethics](https://www.amazon.com/Ethics-Computing-Culture-Advanced-Topics/dp/1111531102)
class, but to say I was familiar with their works would be a lie. I read
**Machiavelli**’s *The Prince* about 10 years ago, and my big takeaway
is that no one uses the adjective “Machiavellian” appropriately. And a
couple of **Aristotle**’s greatest hits have been sitting in my
Goodreads to-read queue for over a decade.

I don’t want to read 52 philosophy texts. But I kindof want to read five
philosophy texts. This quote from the beginning of the video really
stuck out to me:

> Most importantly: in chronological order. That was actually pretty
> important because I realized that philosophy is a conversation.
> Wittgenstein is reacting to Kant, who is reacting to Hume, who is
> reacting to Descartes, who is reacting to Plato, and so on.

The creator helpfully listed all 52 philosophers they read in the
video’s description. But how to reduce that down to just one-tenth of
the repertoire? Well, a neat thing about popular philosophy is a lot of
it is in the public domain, and basically all of it has been analyzed
and discussed intensively. Sounds like a great subject to bring to the
generative pre-trained stochastic parrot!

So I asked ChatGPT 5:

> Let’s say I want to spend a few months really digging into philosophy.
> I want to read a seminal work or educational material from 5
> philosophers over the course of history and I want to do it in
> chronological order so I can see how they are responding to previous
> thought. I want to get a strong sense of the major foundations of
> philosophy. I don’t want to dwell very much on theological philosophy
> but I don’t want to completely ignore its role in the evolution of
> thought. Given all this, and what you know about my interests, can you
> pick 5 philosophers from this list \[from the video\] that I should
> read. Explain your selections. Think of this as my self-guided
> education in philosophy.

It only took a little back-and-forth to land on my list. For largely
arbitrary reasons, I rejected the initial recommendation that included
**Augustine** and swapped in **Descartes**. This left me with a compact
reading plan, with the goal of a crash course on how Western philosophy
keeps answering itself:

1.  **Plato**: *Republic* (Books II–VII)
2.  **Aristotle**: *Nicomachean Ethics* (I–III, V-VII, X)
3.  **René Descartes**: *Meditations on First Philosophy* (I–VI)
4.  **David Hume**: *Enquiry Concerning Human Understanding* (entire)
5.  **Immanuel Kant**: *Prolegomena to Any Future Metaphysics* (entire)

Now let me tell you: if you roll into a Half Price Books with \$40 and
an appetite for pre-owned philosophy, you can shop like a king. I also
picked up some **Nietzsche**, **Smith**, and **Keynes**. Feel free to
yell at me in the comments about whether economics counts as philosophy.

![My philosophy book haul from Half Price
Books](philosophy-book-haul.jpeg)

By now, I’m sure some readers are mad at me. Maybe you work in liberal
arts academia. Maybe you think my tight-five philosophy reading is
deeply flawed and this is further proof that LLMs are bad. That’s okay!
I’m not trying to replace academia. I’m just trying to read some
philosophy without decision paralysis.

I didn’t duplicate it here, but in addition to enumerating works to
read, I went back and forth with ChatGPT to get a bit of a mini syllabus
for insights and themes and speed bumps I might face while reading. It
was a great way to jump in! So I jumped in.

I took *The Republic* one book at a time. My workflow was to read a
book, jot notes in [Bear](https://bear.app/), and then once I finished
paste my messy notes back into ChatGPT for further exploration.

Sometimes my notes were basic, like as I figured out that Plato was
ventriloquizing **Socrates**:

> Hey quick aside: should I be reading these ideas as belonging to
> Socrates or to Plato?

Sometimes my notes teed up my interpretation and projections:

> Man, the noble lie really does make me think about how conservatives
> romanticize the founding fathers and can’t handle the critique of
> slavery

Sometimes my notes were actual critiques:

> It’s pretty funny that this book ends with a grand distinction between
> knowledge of how things “are” vs just having opinions, when all of the
> dialogue thus far across multiple books have leaned entirely on
> enumerating endless things that Plato asserts are “good” or “obvious”
> or “desired”

It would have been fun to do this sort of discussion as a reading group
that might actually argue with me about these notes. ChatGPT
occasionally pushed back, softly suggesting I missed the point, but it
generally told me my observations were sharp and in line. Philosophy is
one of those subjects where you *need* to be challenged more than be
told that you’re right. I’ve probably devastated more liberal arts
professors at this point. That said, I am still the reader, I’m the one
thinking about the works for hours, and it’s my own notes being dumped
into the text box.

So what *are* my thoughts on Plato? Well, a nice benefit of language
models is they’re pretty good at [modeling
language](https://matthodges.com/posts/2025-08-02-language-model-embeddings-campaign-donors/).
They can take twelve days of unstructured commentary and summarize it
all for you quite nicely:

> I would like you to look back at our conversation about Plato’s
> Republic, look at my observations and our discussions of them, and
> highlight my most salient critiques and insights.

- **Epistemic overreach:** Plato asserts a single, static “Form of the
  Good,” downgrades everything else to mere opinion, and demands
  first-principle certainty while rarely supplying it. It’s a bit of
  bait-and-switch: opinions masquerading as knowledge, crowned by Book
  VI’s line without a method for identifying real “knowers.”
- **Static ideal vs. evolving justice:** I think history shows us to
  prefer dynamic, (maybe fallibilist?) frameworks over Plato’s timeless
  blueprint.
- **Authoritarian pedagogy:** Book III’s cultural program: censorship,
  fixed musical modes, and anti-dissonance, reads like “the medium *is*
  the message” turned into policy. Moral psychology by force: control
  styles to control souls.
- **Representation shapes character:** Even though I reject the strong
  censorship, the media-morals link is prescient
  ([Postman](https://matthodges.com/posts/2024-07-20-amused-entirely-to-death/)
  vibes): forms and styles train dispositions.
- **Justice redefined as hierarchy:** In Book IV, justice = each part
  doing its own work (city: rulers/auxiliaries/producers; soul:
  reason/spirit/appetite). It’s a rigid caste system rebranded as
  “harmony,” hostile to mobility and innovation.
- **Noble Lie as civic glue:** The metals myth links to modern origin
  myths (e.g., romanticized Founding Fathers): a stabilizing story that
  resists critique.
- **Elitism about the demos:** Book VI’s ship metaphor doubles as
  anti-democracy *and* anti-relativism. Politicians serving constituents
  chase popularity; truth requires expertise. The case is underargued
  but an anti-relativist thrust is clear.
- **Anti-relativism and anti-novice:** The ship image isn’t without
  merit: some domains are truth-apt and expertise-laden; popularity
  doesn’t make them so. The question is whether “justice” belongs there.
- **Dialogue as ventriloquism:** The “dialogue” is a stacked-deck
  monologue. Plato borrows Socrates’ authority to sell his own system;
  interlocutors mostly assent on cue.
- **Eugenic, anti-private life for guardians:** Book V’s communal
  breeding, child exposure, abolition of family/private property,
  authoritarian eugenics are all very disturbing! Plato pairs it all
  with a surprising argument for women’s political equality.
- **Math as royal road, science half-seen:** Plato exalts mathematics,
  dismissing sensory astronomy, and half-anticipates physics (movement
  in 3-D) while calling current work “absurd.” Admiration for math;
  impatience with empirical practice.
- **Philosopher resentment and duty:** Book VII’s Cave: knowers must
  “return.” Both civic duty and a chip on the shoulder that society
  should revere philosophers more than it does.
- **The big detour problem:** Across the middle books, Plato wanders
  from “is justice better?” into city-design and epistemology. It’s
  engaging but unconvinced: impressive scaffolding, shaky footings.
- **Models vs. methods:** Plato gifts durable *models* (Sun, Line, Cave,
  Gyges, Noble Lie), not rigorous *methods*.
- **Education as moral engineering:** Education isn’t ornament; it’s the
  state’s engine. Book VII’s decades-long curriculum is the hinge
  between epistemology and politics.

This post is not a book report, but rather a report on how I’m using AI
tools to learn new domains in new ways. The tension here is that
philosophy resists tidy summaries, and language models are built to
produce smooth, confident digests. Without bringing your own full
thought forward, they can sand off the strangeness and ambiguity that
makes these texts worth reading in the first place. What I’ve found
useful is not treating the output as my own interpretation, but as
scaffolding: a way to surface themes, keep momentum, and occasionally
highlight blind spots in my notes. The thinking still happens on my side
of the screen; the model just lowers the friction of staying engaged.
I’m going to keep with this process as I work through my philosophy
stack.

On to Aristotle.