I like the idea that essays go places even the author doesn't expect.^[1] Essays are like expeditions where you set out with a destination in mind, or maybe just a general direction. You might never get there, or you might wander down a path in a new direction, because along the way you'll notice something you didn't expect to see. The challenge is to get distracted by genuinely interesting things, and then not too many of them that your writing becomes incoherent. The other challenge is to set yourself off in a direction that's likely to include something of interest.

The latter is easy to do: Just go in the direction of something you're curious about. Even if you end up on a path well-worn and known to others, it will probably be new to you. An essay that reveals insight only to its author is still valuable, at least to its author. And it's a rare path that all but one person in the world has traipsed through.

Another trick I've learned is to write for a specific audience. If you're writing a blog post, pretend instead that you're writing a letter to a friend. This helps in the same way that having a real user in mind helps when writing software: it lets you focus on what matters. It also has the benefit of making your writing more conversational. A lot of writing is bad because it's written in a way that nobody would actually speak.

It's easy to start writing for stupid reasons, like trying to impress people by showing how clever you are. As soon as you let your writing become about that, it's less likely you'll convey the ideas you want to.^[2]

You can also write as though you'll be the only person to ever read what you write. This also helps keep you from trying to sound smart, since there's no chance you'll ever impress anyone with your writing.

Writing for a specific audience and writing for only yourself map nicely to the two primary reason I think writing is valuable. The first is to make an argument or convey some idea or emotion. In this case writing with a friend in mind helps because you'll write more like you would if you were trying to explain something to him. You'll focus on being clear rather than clever.

The second reason to write is to discover something new. In this case you should write for yourself, because knowing that nobody else will see your writing will not only let you forget about style but also make you less inhibited. This means you'll be more free to explore ideas that you're unsure of or that exist in uncomfortable or unconventional places.

1. I'm pretty sure Paul Graham wrote something about this, but I can't find which of his essays it was in.↩
2. I think this is true even if what you're ultimately trying to convey is how smart you are, because odds are your writing isn't exactly about that, so your best strategy is still saying something smart about something else, and letting people draw their own conclusions. ↩