Writer’s Diary #40: Makeshift Writing

Writing is often seen as a pursuit of perfection—the perfect sentence, paragraph, or article. But perfection is the result, not the method. Perfectionism tells us little about the process of writing. Instead, the secret is to embrace imperfection, and approach writing as a makeshift endeavour. It is something cobbled together. Words and ideas are improvised. Perfection, if it is to come, emerges out of an iterative imperfect process. This is the essence of what I think of as makeshift writing. A mode that is temporary, contingent, ever good enough. The point is not to achieve premature perfection, but to work into the words and let the ideas develop, on the page.

Too often, as writers, planning becomes paralyzing. What is the argument? What is the outline? What is the conceptual insight? Research becomes endless. So much so, that it impedes writing. To write this way, is to fall into a trap. The trap is the idea that a piece of writing comes out perfect, and that it must follow a strict, logical series of steps. The goal might be such a piece of writing. But, the getting there is far messier.

The makeshift writing I have in mind is a practical, embodied labor. It happens not in the realm of the perfect analytical structure and theoretical contribution. But in the countless moments of revisions on the page. To write through makeshift is to embrace messiness. A call for makeshift writing is a call to write, to get the words down on the page, and then to improve on them. To iterate. Makeshift is a riposte against the idea that words emerge perfect.

As an anthropologist, maybe one way to approach writing as craftwork is to approach it as an ethnographer might. To think about writing is to embrace an autoethnographic observation of ones practice. It is to consider writing as a process. As a labor practice. To think of writing as labor is to think of it less as intellectual labor, but as a practical, embodied labor of the countless steps and interventions made as a writer. Could focusing on this labor, on the actual practice of writing and fixing verbs and revising tens and making small changes and looking for repeated words and all the other little tricks of a writer’s toolbox help demystify the process?

This process of writing is, for me, ever makeshift. The trick is to begin then to make changes. To be being willing to cobble something imperfect and then improvise and iterate to improve it. To bring together ideas in unexpected ways, and then to work and rework them until they make sense. Makeshift is about approaching the words not as if they were the pursuit of perfection, but to do so as a practical kind of craft work that can be honed through imperfect, iterative hard work. Makeshift writing is a way to get the words on the page, to keep working with them, to keep moving, even in the face of imperfection. It’s about embracing the process, the labor, the craft of writing itself, and it is about how, through this embrace, that something good enough begins to emerge. Something that is meaningful and true, and just maybe, near perfect.

Writer’s Diary #33 – Emergent structure, or work more; think less

This morning, I did a lot of fiddling and reworking the text of a chapter, looking for it order. I read, printed it out, made tentative edits, and reviewed it. I moved chunks of text around, created an outline, and made changes iteratively. After the morning, I have the beginning of an outline.

I can’t see through the project, but I can begin to feel like I could maybe move through it and begin to make sense of it.

What did I do?

I worked at it from 6:00 am to 9:15 am. I read bits of it but moved them around in an outline. I did some queries and a little programming, but mostly I organized, reorganized, worked, and revised. I had no plan. But I read and edited and outlined and moved and renamed. I had a shower, and remembered a scene, and went and found that, and I started to find some order.

Is it perfect? No. But I can begin to get a sense of the chapter.

Tomorrow, I will keep at it.

Mostly, it feels like a structure is emerging. That’s more than I had on the weekend. It feels great.

Writer’s Diary #28 – Anthropic’s Claude-2-100K

Yesterday I listened to the latest Hard Fork episode, which features an interview with Dario Amodei, one of a group of OpenAI employees who left to start Anthropic, OpenAI’s competitor. The interview is quite interesting for someone like me who doesn’t know this area well. But it made me want to check out Anthropic’s new [Claude 2 AI] (https://claude.ai/login).

Claude 2 is not available from Anthropic in Canada, but I got access through Quora’s Poe app.

I’m an anthropologist, a writer, and I’m interested in these big language models and teaching and writing. On the one hand, there is a lot of disruption. I’m on sabbatical, but I don’t know what that means for student essays. But I’m fascinated by AI as an editorial assistant. Many writers rely on editors, but for a long time those editors were unpaid, often female relatives. AI provides me with a free editor and critical reader, for the first time not a family member or friend.

Working with Claude 2-100K yesterday, I came to a couple of realizations.

First, 100K makes Claude much more useful and less stupid than ChatGPT. A few weeks ago I had the idea of using ChatGPT 4 to rearrange a book of fragments. Much of my work involved condensing fragments into summaries for ChatGPT 4’s short memory, or at least the short memory I have access too. With Claude 2-100k, I could give the AI the whole draft. We discussed several possible outlines. With the ChatGPT 4 model I have access to, it always felt like I was at the limit of the AI memory and capacity. With Claude-2-100k, the answer was much slower to come, but was feedback on the entire damn manuscript. My first book took 8 months. This took under a minute.

From there, I had a long, free-flowing conversation with Claude 2 about my book, its ideas and theories, and then we turned to writing and the differences between how Claude 2 writes and how I work. Frankly, Claude 2 just felt much more thoughtful and interesting to talk to than ChatGPT 4.

ChatGPT feels like it has less memory, is dumber, and frankly more prone to nonsense. It seems smart at first, but on further review it is often very formulaic.

Maybe Claude 2 will seem that way to me, but yesterday it seemed like a much more interesting interlocutor than I’ve had on this book, ever. What it says about me that I say this about a large neural net running on servers, converting text into tokens and then running complex vectors and matrices, leaves me very confused. What does it mean to be human, I wonder?