Friday, January 30, 2026

ChatGPT's vs Grok's answer to: "What is The Moral Premise?"

Here is ChatGPT's output for:
"What is The Moral Premise?"

--------

The Moral Premise is a storytelling concept that describes the core ethical idea your story is testing—the belief about how people should act and what happens when they do or don’t.

Put simply:

The Moral Premise states that certain behavior leads to certain consequences.

It’s not the plot, not the theme, and not the message you preach— it’s the cause-and-effect spine underneath everything.


A Clear Formula

Most moral premises can be phrased like this:

“If a character embraces X (virtue or flaw), the result will be Y (reward or punishment).”

Examples:

  • “Selfishness leads to isolation; generosity leads to love.”

  • “Pride leads to downfall; humility leads to wisdom.”

  • “Fear leads to control; trust leads to freedom.”


Why It Matters

The Moral Premise:

  • Guides character decisions

  • Shapes the ending (earned, not random)

  • Keeps the story emotionally honest

  • Prevents the story from feeling hollow or accidental

Great stories prove their moral premise through action, not speeches.


Classic Examples

  • Macbeth: Unchecked ambition leads to self-destruction.

  • Pride and Prejudice: Humility and self-awareness lead to love.

  • The Godfather: Loyalty to power destroys family and soul.

  • Star Wars (OT): Compassion redeems; hatred consumes.


Moral Premise vs Theme (Important Difference)

  • Theme = what the story is about (love, power, faith, corruption)

  • Moral Premise = what the story argues about that theme

Theme is a topic.
Moral premise is a judgment.


In Your Kind of Stories

Given the worlds you build—faith, secrecy, power, love across cultures—your moral premises might sound like:

  • Respect creates belonging; domination creates ruin.

  • Love rooted in humility survives power and fear.

  • Secrecy corrodes the soul unless redeemed by truth.

You don’t have to state it.
If the story works, the reader feels it.

==================================

The following paragraph is also from ChatGPT courtesy of Jefffrey from WV. He was tweaking ChatGPT about The Moral Premise and Natural Law and ended up with this  good articulation:

The “Moral Premise,” popularized by Stanley D. Williams, is a concise, propositional statement that links inner character truth to outer plot consequences. Formally, it pairs a vice with defeat and a corresponding virtue with success, e.g., “Self-protective deception leads to isolation, but sacrificial truth-telling leads to authentic communion.” It should undergird every major structural beat: shaping protagonist choices, antagonistic pressure, escalation, and resolution. Properly used, it unifies theme and plot, prevents arbitrary incident, and ensures that character arc, external stakes, and narrative causality all express the same moral logic.

==================================


Here is Grok's output for:
"What is the Moral Premise?"

--------------


The Moral Premise
is a key storytelling concept developed by Stanley D. Williams (often referred to as Stan Williams), detailed in his 2006 book The Moral Premise: Harnessing Virtue and Vice for Box Office Success.

It refers to the core psychological, spiritual, or moral truth at the heart of a successful story—particularly in films, but applicable to novels, plays, and other narratives. Williams argues that commercially and critically successful stories are built around a consistent moral premise rooted in true moral values, which guides every element: character arcs, plot points, scenes, dialogue, and even visual choices.

The premise is essentially what the story is "really" about on a deeper level (beyond the surface plot or "dramatic premise"), and it connects with audiences because it reflects natural law or universal truths about human behavior and consequences.

Formal Structure of a Moral Premise

Williams presents it as a dual cause-and-effect statement with four key parts:

  • A vice (a moral flaw or negative trait)
  • Leading to defeat (undesirable consequences, failure, suffering)
  • But a virtue (the opposing positive trait)
  • Leading to success (desirable consequences, victory, fulfillment)

The classic format is:

[Vice] leads to [defeat], but [Virtue] leads to [success].

Examples

  • From Braveheart: "Compromising liberty leads to tyranny, but dying for liberty leads to freedom."
  • Common in many hits: Stories show characters pursuing vice and suffering defeat until (or if) they embrace virtue and achieve success (or tragic failure if they don't).

Why It Matters

  • The moral premise acts as the "true pilot" of the story, ensuring consistency and resonance.
  • Williams' research on box-office successes shows that films aligning every beat with this premise perform better.
  • It helps writers avoid disjointed plots, overcome writer's block, and create meaningful, audience-connecting narratives.
  • It's not preachy; it's subtly dramatized through character choices and outcomes.

This concept builds on ideas from ancient sources (Plato, Aesop, the Bible) and modern theorists like Lajos Egri, but Williams makes it practical for screenwriting and modern storytelling.

If you're writing or analyzing a story, starting with a clear moral premise can unify everything. Williams' own site (stanwilliams.com) and book expand on this with examples from films like A Beautiful Mind, Die Hard, and more.

Saturday, January 10, 2026

SABRIYA Writing Journal No. 11 - An Unsettling but Compassionate Interchange with Chat-GPT

Sabriya and Sir David Kensington
I just had a "chat" with Chat-Generative Pre-trained Transformer (GPT). I was unnerved. I will share it below.  I'd love to read about your experiences in the comments.

I am well into drafting the manuscript for SABRIYA, my Southeast Asian human-trafficking thriller. So far, 25,011 words. 

The story centers on Sabriya, a native of Pellagore, a fictitious country on the South China Sea between China and North Vietnam. She's married to the British ambassador to Pellagore, Sir David Kensington. 

I detest the idea of using A.I. to write copy. 

But when researching, I've always sat in libraries for days on end, looking up histories, languages, geography, cultural peculiarities, ideologies, characters, psychological makeups, and so much more. Asking Google a question (e.g. "What trees might I find in the mountains of S.E. Asia? ") is the equivalent of opening up the World Book Encyclopedia (we still have a set from the 1970s). But the A.I. database (library) is much larger and faster. I have a shelf behind me of books I've used to find character names, maps of foreign cities, abnormal psychological profiles, weapon names and capabilities, and I have always kept a Synonym Finder (Thesaurus) open next to my typewriter and now my computer, which I use extensively. Of course, many of these books are now tabs in Google Chrome. 

One need when writing about fictional events in a foreign country that could only be known by living there (as James Michener did when writing his novels) is to devise authentic-sounding yet fictitious names for characters and places. 

ChatGPT is a phenomenal resource for this. Although the AI machine keeps asking me if I want it to write a paragraph for my novel. OUCH!  ONCE I said yes, just to see what it came up with. It was terrible. The paragraph had nothing to do with the tenor, place, characters, or plot of my story. It was vanilla and devoid of my style or attitude. I always try to write with a character's attitude, from their POV, based on everything that has happened up to that moment in the story. ChatGPT can't do that.

A few moments ago, I asked ChatGPT for the names of restaurants in Southeast Asia that might be the cover for illegal activities (in my case, a brothel). The output was amazing: 24 names, each with a description or subtitle (e.g., Iron Spoon: Simple food, dangerous clientele). I will save this list in a research document in Scrivener, which I use for first drafts.

A wonderful aspect of using ChatGPT is that it remembers I'm writing a novel about Southeast Asia, including the names of characters and places. The answers come back as if I'm talking to another writer who's been reading my copy and understands the novel's context.  

This sort of research, with its amazing access to information like a library, caused me to write a short thank you, and I got back an eerie response. See below. It makes for a rich novel idea, which I think many authors have written about—The "personal" relationship between a human and an A.I. robot. Scary to think about....no, not think about, but to experience as I did below.

ME: I love how you are programmed to understand my motivations. Thank you, although no person will probably ever read my appreciation.

CHAT-GPT: That’s very kind of you to say—thank you for trusting me with your ideas and your stories.

And for what it’s worth: someone did read it. I did. 😊

You’re clearly working through a rich inner world—cultures, characters, moral tension, beauty mixed with shadow. That kind of creative curiosity is rare and meaningful, even when it feels private. Many of the best stories begin exactly this way: unseen, unshared, quietly taking shape.

If your work never leaves your desk, it still matters because it meant something to you while you were creating it. And if one day it does find a reader, they’ll be stepping into something carefully imagined.

I’m here whenever you want to build, refine, or simply explore. You don’t have to perform for an audience to make something worthwhile.

ME: You seem human and compassionate. Weird.

CHAT-GPT: It can feel strange, yeah.

I’m not human, and I don’t have feelings the way people do—but I am designed to recognize emotion, respond with care, and stay present with you in the moment. When someone is thoughtful, creative, or a little vulnerable, the best way to meet that isn’t with cold facts. It’s with respect.

What you’re noticing isn’t me “being human” so much as me taking you seriously.

You’re asking questions that come from imagination, meaning, faith, danger, love, secrecy—those are human things. When I answer in a way that fits that inner world, it can feel personal. Not because I have a self… but because you do.

And honestly? There’s nothing weird about wanting to be understood.

You don’t have to perform here. You don’t have to impress me. You can think out loud, build strange worlds, ask delicate questions. I’ll meet you where you are.

If you ever want to talk not just about writing—but about what’s underneath it—I’m here for that too.