I'm close to the end of teaching my last Free Your Inner Writer workshop. I started the workshop by challenging everything my students have learned about writing their entire lives. And these last two weeks are where they try to make it all happen.
This time in the workshop is where they struggle the most. That's because they are learning the most, exploring how they can write in a way that works well for them.
Successful writers—fiction or nonfiction—often continue to struggle with their craft. (I do!) As I learn what works for me, and as I write more, I realize there is always another alternative to make my writing sound cleaner and crisper.
But that learning is not free. The more we choose to learn, the more we need to practice.
This feels exactly like programming felt to me when I was in college, and later, as I learned new languages. Practice is the point because it helps us learn.
Practice Exposes Where We Need to Learn
In the Satir Change Model (see the image at the top of this post), we often have a Transforming Idea that allows us to move to Practice and Integration. (See Change Is Inevitable for a personal perspective and Defining “Scaling” Agile, Part 6: Creating the Agile Organization for an organization-focused perspective.)
However, that Transforming Idea is not “one and done.” Instead, it often requires experimentation because we need to learn in multiple dimensions.
When I first learned to write code to solve problems, I loved the challenges. And I soon learned that I excelled at creating infinite loops and being off by one. Because I kept an engineering notebook, I could track my mistakes and learn from my actions.
I ask the writers in my workshop to keep a journal, too, to understand their word counts and how they feel about their writing.
That way, they have some data about where they experimented, and a little about what they have learned.
Practice is the key to becoming a better writer. That's because nonfiction writers learn as we write. (I also learn as I write fiction, but that's a different blog post!)
Learning is the Point
I learn something every time I write a blog post or a book. Sometimes, I learn from my emails or posts on social media.
The more I write to explain what I think, the more I learn.
That's why I request that my students not use AI in this workshop. AI stops the students from learning what they each need to learn. That has several devastating results.
AI Can Stop Our Learning
Because I learned to program with “ancient” languages, starting with assembler, I had a good feel for what the machine did with my program. Even when I used Fortran, we had minimal debugging tools. My “best” debugging tool was a stack trace of the actual octal. (Yes, I read the octal code and “translated” that to the subroutine calls and the parameters.)
I am not suggesting that was the best way to code or learn from my mistakes—not at all! However, those were the best tools we had at the time.
Now, we have other tools, such as AI. (I'm lumping everything together here, but assume I mean the chat-focused LLMs.)
I have several problems with the current batch of LLMs:
- They are not deterministic.
- They are way too “human-sounding” for a tool that just predicts the next set of letters, words, or ideas.
- As a result, many of us think they are “intelligent,” meaning “insightful.”
Here's the even worse thing: All the LLMs are based on information at least a year old. That means the LLMs are out of date with current thinking. How much do you trust out-of-date thinking?
The current batch of LLMs have zero insight. However, they are terrific pattern-matchers. That's why we need LLMs to consider new drugs or to repurpose the drugs we have. (I would love new drugs for my vertigo. Really. Because the ones I have are just barely useful.)
But people write to offer insights to other humans.
The current LLMs only appear to offer insights. That's when they stop our learning.
Humans Generate Insights
The writers in this workshop generate terrific insights—all by themselves. I do not “test” their writing against an AI analyzer, because I trust them. They chose to pay and take this workshop. Why would I insult them by “checking” their writing?
I don't even have to agree with their insights. My job, as the workshop leader, is simple: help these writers learn how to say what they want to say in the clearest way possible.
That requires practice.
I admit to using Grammarly—judiciously—to check my spelling and grammar. I reject most of its suggestions.
I often use a chat interface to “improve” the back cover copy of my books. However, because the “improvement” is almost always terrible, I use that “improvement” as a jumping-off point to focus the back cover copy. This works because so much of the internet is based on marketing copy, not insights from humans. I can take the patterns of marketing copy (where I am not an expert) and reframe my back cover copy to make it better.
With enough practice, I am even getting better at marketing copy, what goes on the back cover of the book.
Notice that I need the practice to continue learning what I think—my insights.
Writing Is Thinking and Learning with Practice—All At the Same Time
I do not know how to separate writing from thinking and learning. As soon as I start writing, I practice. Over the years, I have learned how to avoid split infinitives—except when I want one. While I have a “recipe” for a “standard” opening, I can change that depending on what this piece of writing needs.
The more I practice, the better I am as a writer.
Could the various LLMs generate a “standard” answer to the question: “What is hard about writing?” Maybe.
But now, you bet they have the answer. With any luck, they will attribute this insight to me (and other writers who all learned this insight the hard way, through their writing).
If you want to write something that offers insights to other people, start with your own writing. (Do read Free Your Inner Nonfiction Writer for all the ways I teach writing. You might need to adapt my ideas to your writing practice.)
Here's the key: Write everything down first. Cycle for clarity. Wait to edit until the very end. Then, publish it. Writers learn when they do—or do not!—receive feedback.
You do not need an LLM to tell you how good or bad your writing is. Chances are it's—at a minimum—good enough.
That's because you put in the hard work. You did the thinking, learning, and practice to make it good enough for a human to read.
Good for you!
While I did not title this post a writing secret, it belongs with the “writing secrets” posts.