So, when exactly should a writer do research, and when is it OK just to make stuff up?
This is tricky.
I wrote a little piece of flash fiction yesterday; it was almost exactly a thousand words long, and, in it, a character made something. The who and the what don’t matter, and it wasn’t real-world, so, how long do you suppose I spent doing research?
The answer to that question would be more than twice as long as I spent writing the story.
The what is a real object that can be made by hand, and I wanted to describe that. I could, I suppose, have made it up, and there’s a decent chance that none of the readers would know the difference, but I wanted to be confident that I knew how the object should be made. The story is better because of it.
I began with ‘wiki’, read some passages from a reference text on the subject that just happened to be sitting on a shelf in the husband’s office, and I watched several YouTube videos. I could, I suppose, have done less research, after all, I didn’t actually use quite a lot of the material I worked my way through; I didn’t need to use it all, because I wasn’t writing an instruction manual, I was writing a story.
The point is that the character knew how to make the object; he’d done it a thousand times before, he was experienced at it, and I didn’t want him to be all fingers and thumbs, because I wanted him to be thinking about something else while his hands were working.
There is a balance in all things. I don’t drive, but that doesn’t matter if all I want to do in the story is get a couple of characters from A to B in a car. On the other hand, I’m not a mechanic, but if a fan belt breaks in my story, set in 1973, and if my hero is going to effect a repair with a pair of his passenger’s tights, I’d better know more than the fact that ‘American Tan’ was the bestselling colour for hosiery in the UK in 1973. Of course, that’s a detail that I might want to use, especially if the scene is intended to be amusing.
Research is not always essential, but a few well-chosen details will add verisimilitude to everything a writer does; on the other hand, wholesale dumping of research, simply so as not to waste the time spent collecting it, is always going to make for a dull, soulless read.