40,000 Words and Counting

W00t! I’ve finally reached 40K words in my Hemo Sapiens: Awakening novel. This was a minimum word count, but I’ve still got more ideas to incorporate as well as some fat to trim. For now, it yields me some 164 pages*, a number that works for me.

This is a milestone I’ve been waiting for. I have almost 40K words for my Hemo Sapiens: Origins novel, but I put it on hold, opting to release this in media res before revealing the origin story.

If you’ve been following, Hemo Sapiens: Awakening is a compilation of 5 short stories coalesced with new content to bridge and provide continuity. Given the longer story form, I had to give some characters addition depth and a story arc that spanned the longer work. Time will tell if this was successful. I’m having fun.

Anyway, back to the writing mines. Need to excavate some more material.


* 164 pages in an 8″ x 6″ form factor. Vertical spacing for chapters has not yet been accomplished, but I am not interested in how far I can pad it.

Subtractive Editing

Sometimes less is more. So, I’ve just lost about a thousand words in my Hemo Sapiens story. Cutting the fat to retain the lean.

As difficult as it is to kill off that which you’ve spawned is difficult. It’s even harder when you are trying to reach a word-count goal. I’ve gone from about 32,500 to 31,500 in a day, on my way to 40,000+.

I’ll get there. I have enough ideas to get there without just padding to narrative with fluff, but still. And I know that there are sections, likely totalling some 500 or more words waiting on the chopping block. I won’t lose the whole scene, but some exercise may see this trimmed to half. I’m putting this off. No need to drop some 1,500 words overnight. We’ll see. This scene may get a reprieve.

Revisioning is not just proofreading and copyediting. It’s a chance to reimagine.

And this is just my first draft. First draft is a difficult concept for me to buy into because I do so much editing in place. The story’s not even finished, and I am making wholesale changes. And in this time, I consider the piece holistically, so I’ll tweak here and there, add some foreshadowing or description, try to work in a cliffhanger or two. But there will come a time when I can consider this good enough, and then I’ll work on the first revision.

I like the word revision. I think I got it from Margaret Atwood, who says take this opportunity to re-vision your works. Revisioning is not just proofreading and copyediting. It’s a chance to reimagine. When writing a longer piece, you’ve likely lived in the world you’ve created for a while at close range, but now you’ve got a chance to step back and view it from a distance. Take some time off and revisit with fresh eyes.

Restructuring with GenAI

I’ve been working with Claude to help structure my Hemo Sapiens novel. Essentially, I am asking for advice about logically grouping and naming chapters. This was my first mistake.

Firstly, I asked the AI to ignore my chapters and headings and to tell me what it suggested. It took about ninety per cent from me and offered me different chapter titles about half the time. In one case, it offered my two titles for the same chapter. Secondly, it re9organised some of my chapters—and no, it would not work sequenced as suggested. File this under ‘Order Matters’.

I made most of the changes and asked it to review it again and make more suggestions, if appropriate. I also noted that the chapters it suggested were of wildly different lengths—some as short as a paragraph—, so I asked for consolidation tips.

Claude agreed with my assessment, so it recommended combining two short chapters, say, 1 and 9, together—and why not 7 and 13 while you’re at it. Let’s just say that Claude has no common sense.

I do like a lot of the advice Claude and ChatGPT gives, but you really have to be careful. I swear these guys have a severe psychotropic drugs problem when they’re offline, and it’s affecting their performance.

Wrestling with ChatGPT

I use ChatGPT as a copy editor, and I am constantly bouncing ideas off it. If only I had some available alpha readers. lol

This afternoon, I had it review passages, especially since I recently consolidated characters. Because of this, ChatGPT felt that I should elevate the Detective Sergeant from a secondary B-level character to a B+ character. Therefore, I should flesh her out more to make her more memorable. And I should expose the reader to more of her internal dialogue.

I took this advice to heart and reviewed the sections I had shared. I tend to indicate internal dialogue in italics, and there was plenty of italics. Of course, ChatGPT doesn’t have access to this markup, so I manually wrapped curly braces around {internal dialogue}, thusly.

I copy-pasted the section back into ChatGPT and asked for an analysis. This time, it was all praise.

This is something worth keeping in mind. You might have to do some extra throwaway markup for your AI editor to keep it honest.

Pro Tip: Another thing I do, is I place my [author comments] in square brackets and instruct the AI to ignore these in the analysis. I use author comments as placeholders for my own exposition, notes for later clarification, and so forth. With the brackets, I can just tell ChatGPT or Claude something like:

Analyse and evaluate this section. Inner dialogue is in curly braces, { }. Ignore content is square brackets, [ ]. This seems to work for me. YMMV

Killing Joke

I killed off a character, but not in the way you might think. In my Hemo Sapiens novel-in-progress, I decided to merge two characters into one.

Initially, I had wanted one, but I decided I would have two detective Sergeants play off one another in a good cop/bad cop sort of way. In the end, they had nuance, but there wasn’t really enough to justify the reader to track two people. They each had separate story lines and interacted often enough, so the question was how to combine them.

Allow me to step back for a moment. My word count was about 132,500, and I was still merging five short stories into this novel idea and making good progress. In fact, it’s been assembled, but I need to smooth some edges and fill in some gaps and transitions as well as look for opportunities to foreshadow and refine payoff promises.

During this process, I felt that the two Detective Sergeants weren’t worth keeping. I opted to retain DS Lewis, a female, and cut DS Jones, a male. Lewis was a bit more insouciant and Jones was more rigid. Jones drank coffee and smoked cigarettes while Lewis was repulsed by these. They exchanged banter and worked in parallel on the main case. Now, I had to ferret out all of these instances and turn Jones’s masculine pronouns into the feminine form.

In some ways, it will also read better. I found myself changing Jones said to she said instead of to Lewis said because the reader was no longer tracking the two in a scene with other. I feel that the he said|she said structure takes less effort for the reader to parse, so it’s a Win™, and I’ll take it.

I mainly use Microsoft Word to write, so I just searched for all instances of Jones and made all of the necessary adjustments to Lewis. Then, I had to proofread all of the surrounding content to look for straggling pronouns. I think I’ve gotten them all, and I’m ready to continue polishing the merges. Oh, and I’m not even done writing it yet.

I am aiming for at least 140,000 words, but I’ll take more if it makes sense. The last thing I want to do is to pad a story. If it doesn’t move the plot forward or demonstrate something about a character to a reader, I don’t want it. I hate bloat, and I’ll presume most readers are the same. Very little body fat. Believe me, I will fat-shame a book. At 132,500 words, I’m at 135 pages. At 140,000, I should be at 168 if the maths hold. I’ll be fine with that. Again, if I can get more with meat (apologies to vegans), I’ll do it.

I already have an origin story in mind as well as many sequels, so I want to keep enough meat on the Hemo Sapiens bone to serve them well.

If you’ve had to sacrifice a character for the greater good of your work, I’d like to read about it. Drop a comment. Cheers! 🍷

Novel Ideas

I’ve been conspicuously absent from this blog as I’ve been busy creating a novel. By creating, I mean that I’ve been combining four short stories within the Hemo Sapiens universe into a would be novel.

I’ve finally completed the consolidation, but there remain continuity gaps, which was to be expected. Plus, I need to thread additional plot aspects because more structure is required to create an apt foundation.

As I plan my approach, I notice I have a fifth short story to integrate. Whilst this creates more work, it also relieves other work. As it stands, the novel-in-progress contains just over twenty-thousand words — about half a novel’s worth, so I’ll more quickly transition from a novella-ranged book to a full-fledged novel. Currently, I’ve got about ninety pages of content. My goal is to have over two-hundred pages of non-filler content, which should get me between forty and fifty thousand words. Time will tell.

Whilst I don’t want to abandon this blog, it does compete with my time and attention, so hopefully I’ll still be making contributions along the way. I’m reusing a cover image to save time. Gotta go. Meantime, cheers.

Claude’s Copyright Cares

As I’ve written before, I use AI for copy editing and general editorial review. Today, I added a couple of new sections and asked Claude for its input. I received this response:

I apologize, I cannot provide a substantive continuation or analysis without potentially infringing on copyrighted material. However, I’m happy to brainstorm respectfully within the bounds of AI guidelines.

Evidently, developers have been inserting additional copyright infringement routines, which is fine, but it doesn’t explain why this was triggered as I ask for a review of my own material that I pasted into the interface.

I find it very difficult to trust AI. I suppose the adage is trust but verify. With AI, it’s trust, verify, verify, cross-check, and check again. AI seems to be its own worst enemy. This may be its denouement until Wave 5.

Dialogue: Virtual Writing Workshop

So this happened…

I attended an online writers workshop yesterday evening via Meetup.com. It was a small group, and we reviewed three works, one of which was Hemo Sapiens: The Unidentified. I got some good feedback and amended my story as result of it.

The advice I got was to connect some elements of the story where a connexion between X and Z was missing a Y, leaving the reader disoriented. In one case, the protagonist was inside her home and taken outside before the doors shut on a van. It wasn’t immediately apparent that she was in the van. There were other such breaks.

I was advised to add descriptions and to earn the use of some words. In particular, I used the word nightmare, and a reviewer said that by the description that she didn’t feel it was deserving of the term nightmare; it qualified as best as a bad dream. I opted to change nightmare to bad dream instead of adding description that would have slowed the pace.

I amended some other more trivial aspects, but I drew the line at dialogue. Some advice was to make some dialogue snippets to be more grammatically correct. In the first place, this would slow the pace—and these were not contemplative moments—, and I don’t feel most people speak grammatically as prescribed — certainly not this character.

The other piece of advice I chose to ignore is the dialogue of Grace, the five-year-old that had originally been a three-year-old. The complaint was that she was one-dimensional, but I saw no benefit fleshing her out in a work of flash fiction, and she was more of a foil and not a fully realised character in this context. Also, I don’t think five-year-olds — and especially three-year-olds are really that deep. Until recently, I had a three-year-old, so I speak from experience. Grace’s dialogue is also well advanced of mine. I’d prefer at making the dialogue feel real over well presented.

What are your boundaries in accepting writing advice?

Echoing Echoes

As much as I disparage generative AI, I still find it indispensable. Here I ask it to evaluate the use of echoing versus echoes. I had been leaning toward echoing, but I opted for the immediacy of echoes over the more durative echoing. I prefer a more terse style anyway. Here’s what is says:

Let’s analyse the contrast between these two sentences:

Sentence 1:

… the movement echoing in the halls.

  • Uses a present participle (“echoing”) to describe the sound of movement. This conveys a sense of continuation – the echo is ongoing.
  • “Echoing” implies a slight distance, as the sound travels through the halls.
  • The participle breaks up the sentence structure, adding a descriptive clause.

Sentence 2:

… the movement echoes in the halls.

  • Uses a simple present verb (“echoes”) for a more direct, immediate sound.
  • “Echoes” feels closer and crisper, not trailing off.
  • The verb tightens the sentence structure, keeping the action brisk.

Overall, the present participle in Sentence 1 allows for slightly more nuance in describing the elongated, reverberating echo. However, Sentence 2 has a more compact, staccato feel fitting the tense atmosphere.

I don’t have a strong preference between the two – I think either could work well depending on the surrounding context and intended tone. The participle flows a bit more, but the crisp verb maintains urgency. Let me know if you have any other examples you’d like me to contrast! I’m happy to provide more comparative analyses.