This project is about my attempt to make more expressive, unconventional portraits. I’ll show some of my earlier, less successful efforts that got me thinking differently, along with work that’s a bit closer to the look I’m after.

I’ve been interested in portraiture since I started photography, but I quickly found that I don’t enjoy taking ‘straight’ studio portraits. I’m far more drawn to the kind of expressive portraits made by photographers like Sarah Moon, Francesca Woodman and Victor Skrebneski, where the subject is blurred and the viewer is left to do some work interpreting the image. (I especially like Victor Skrebneski’s portrait of Dennis Hopper).

I think I’m drawn to this style because it’s an antidote to how I usually work. My normal approach is precise, controlled, even a little… buttoned-up. My horizons are always level. My focus is razor sharp. My exposure captures the full dynamic range. The problem with this way of working is that it tends to prioritise technical perfection over creative risk. Now and again, perched on my shoulder, I hear a laid back surfer type whispering: “You’re squeezing it too tight, dude. Let the camera dance a bit.”

I feel like a cook who always follows the recipe and weighs every ingredient. Sometimes, I just want to ignore the recipe, trust my instincts, and see what happens when I chuck it all in the pan.

I’ve played with expressive portraiture before. I’ve created 360-degree portraits by blending a series of shots taken as the subject and I moved in a circle. I’ve distressed printed portraits by re-photographing them through water. I’ve cut up printed portraits into strips and woven them together. I've taken 30-second portraits with a film pin hole camera. I’ve also explored alternative printing processes, like cyanotypes. This isn’t a passing interest but something I keep returning to.

Here are three early experiments that didn’t quite work, but got me thinking differently.

Taking portraits is hard when you don’t have easy access to models. I know you can hire a model, but I thought it would be easier to attend a model-day event. These events typically give you access to three models in a two hour session.

I knew that over the course of the day, another 26 photographers would be photographing the same models, under the same lighting, with the same props. If I went with the flow, the odds were slim that I would leave with something original. I would probably end up with the organisers’ view of what ‘good’ looks like: well lit, club competition-ready pictures — but not what I was after. This made it a good opportunity to experiment with some of the expressive ideas I’ve been considering. If nothing else, I'd come back with images that no-one else had taken.

So I planned a different approach. I brought a Lensbaby with me (a manual-focus lens designed to create dreamy, selective-focus effects, where only a small part of the image is sharp and the rest falls away into a blur). I hoped to try some unusual crops and shoot just parts of the body instead of full-length portraits. I brought a prism to hold in front of the lens to emulate light leaks.

But I ran into problems immediately. I liked the way the Lensbaby and the prism both added randomness to the image making process, but nine out of ten shots were unusable. The Lensbaby is tricky to focus, and the prism is equally fiddly. That would have been fine if I’d had time to experiment, reflect, and refine. But I was under time pressure. In my session, there were nine photographers split into groups of three, with each group assigned one model for about 45 minutes. Then each group moves onto the next model. Because we had to take turns, I only had about 15 minutes with each model. I needed more time to play. I wanted to shoot an image that was almost what I had in mind, then tweak the lighting, then adjust the framing, then riff off of that idea and try something different. But the clock was ticking.

Here are two images from the day that hint at what I was after. I’m sharing them here knowing that they aren’t good enough. “Well done for trying something different,” says the laid back surfer on my shoulder. But I know what I tried that day didn’t work.

The model day hadn’t been a success for me so, like all photographers when their images don’t work, I decided to fix things in Photoshop.

I selected a couple of the conventional portraits I’d made at the model day along with some others I’d taken in the past. I then distorted them using Photoshop’s Path Blur filter to see if I could get closer to the look I was after.

It took me a while to understand how I could bully Photoshop to get the expressive feel I wanted. But ultimately I ended up with a set of six images that I'm happy with.

What surprised me about using Photoshop to create these portraits is that it introduced some randomness into the creative process. Each portrait reacted differently to the treatment. It’s a little like in-camera movement or creating double exposures. You’re never really sure about what you are going to get until you’ve finished.

And at the same time the process allowed the buttoned-up version of me to take time, experiment, and make it a bit more perfectly imperfect.

These six portraits aren’t the final word, but they’re a step forward. I think the surfer dude would approve.

If you have any comments on the portraits, good or bad, let me know at david@davidtravisphotography.com.