A Relaxed Approach to Philosophy

A lot of photography revolves around philosophy and attitude. Over the years, discussing with friends and on online forums, my approach has shifted.


My chosen genre is landscape photography, based in Scotland. My inclination is to make more than just documentary photographs: rather than take photos "of a mountain", I prefer to make photographs where some aspect is worth studying whilst retaining a clear link back to realism; hopefully my photographs have elements of visual desirability but with overtones of artistic merit and message (political or environmental) mixed in.

My primary criterion for "artistic merit" is therefore not that a particular subject is well depicted, but rather, that the photographic frame is populated with a subset of the real world in an aesthetically pleasing, balanced, fashion - a photograph is not about the subject, it is about the frame one brings to it. Therefore one of my key understandings of art is that an image arises from a conscious, thought-out, intentional process. A photograph that looks deliberately made is more pleasing than a snapshot thrown together.


One of the characteristic features of a photographic philosophy with which I am most likely to disagree is the presentation of an aspect as absolute when it's really open to artistic choice.

What is "good" light in a landscape?

The normal advice to avoid a few hours around the middle of the day and take photos only at sunrise and sunset in golden light is rife. Frequently this light - especially when it forms pleasant dappled contrasty patterns of light and shadow - is known as "good light".

Similarly, a handful of photographers I know are fond of dull lighting, a blending of grad-ND filters and large-format Fuji Velvia on grey days hiding from the rain, in the name of atmosphere; or hiding around shadows to make photographs in the cold blue light of a sunny day.

Seemingly neither of these can be an absolute value; it does not make sense to only go out seeking or making landscape photographs at one or other time of day - neither only golden hours, nor only in the rain.

With some sub-genres - typically "intimate" or closeup landscape work - it is reasonable that the light should not be a distraction if the photographer wishes to portray an intrinsic property of the scenery in front of them. For example, if the aim of a photograph is to portray the shapes and outlines of a leaf, in black and white, then patterns of sunlight and shadow could wreck highlights and interfere with natural blotches on the leaf, destroying the intended effect. In some ways, this is "product" photography - diffuse even light to illuminate what's there, only.

Rather, therefore, the common factor is to learn to recognise a goodness in light that is appropriate to the scene already in one's mind. This does not mean that either brightness or dullness is either good or bad - too specific, too absolute - but rather, a photographer should walk happily over the earth analysing light for where it fits a photograph.

"Manual is better"

Frequently one sees advice that other people have had an epiphany by switching their cameras from some kind of "Auto" or programme mode to manual mode, normally presented with an encouragement to go try Manual mode yourself.

This is actually getting a fundamental feature of photography backwards.

Yes, at the end of the day, there are only a handful of parameters that control a photograph's appearance - the interplay of ISO sensitivity, aperture (f/stop), exposure duration (shutter speed) are all well-known; secondary parameters such as the sensor's response curves are also understandble. We know that choosing a "landscape" preset makes the camera favour greater depth of field (narrow apertures) and choose a response curve with greater saturation especially in the green and blue channels.

What Manual mode achieves is to express the choices and compromises of all the other preset modes in a flattened lowest-common-denominator form. Rather than choosing both the sensor response *and* an aperture for depth of field, you can choose "landscape".

It certainly pays to know how all these parameters work. Anyone who's used large-format film cameras will know that a lot of the dancing around the lens double- and triple-checking the settings prior to making an exposure is precisely that common denominator - where the sensor response is a matter of choice of film.

What I dispute is that any one mode is better to live in all (or most of) the time than any other.

On the one hand, you might practice the Zone System and choose to place specific tonalities at given output densities, in order to minimise the amount of change required in processing. On the other hand, you might know that when a scene has a roughly Gaussian histogram, by centering a general exposure one maximises the representation of scene-contrast in the output response-curve - the histogram is as wide as it can be when it's centred - which gives greater room for manipulating tonality in post-processing.

Automatic modes work by centering some function of the scene as a midtone-grey exposure. If you know the camera can cope with the contrast in a scene, just point it in the right direction and it will probably choose the best exposure it can, for you.

It is up to the individual photographer to choose what parameters are important. Someone making closeup photos of plants might choose to fix the aperture at f/5 as a combination of lens-sharpness and narrow depth of field enough to separate a flower-head from the background whilst retaining detail within the flower itself; someone choosing wider landscape vistas might choose to fix it at f/11 or even f/16; a sports photographer might decide that having a shutter-speed of 1/500s or even faster is the most important thing to them and let the aperture do what it will. If a photographer is comfortable knowing the camera can expose their scenes well and fast enough and can combine that with imposing their requirements, there is no need to insist on manual mode specifically.

As an example, some newer dSLR cameras have a "Creative Auto" mode, in which one chooses a desired relative exposure (dark or light overall) and a degree of background blur, and the camera chooses an automatic exposure with those features in mind. This is perhaps the best way an artistic vision should be imposed on a scene: we might not need to know what specific aperture value achieves the desired depth of field in a particular scene, but we do know what DoF effect we want, so let the camera work it out.

The role of gear-acquisition in inspiration

One frequently hears two related ideas:

These are not quite right.

The first case fails because it is not the artist that is subject to the machine, but the machine to the artist. If a photographer has a desire, say, to produce acceptably sharp 25x20" prints, then one either needs to start from medium- or large-format film or acquire a dedicated film-scanner for 35mm. If shooting digitally, merely to build a reasonable number of pixels for this job is going to require either an expensive dSLR or medium-format digital back, or some form of panorama-stitching and/or up-scaling from multiple smaller photos - and the closer one starts to the required resolution, the easier it is to achieve. Not all cameras are created equal for all tasks!

The second case has a lot to commend it, on first glance: because the artistic merit is entirely a feature of the photographer, changing equipment is a very poor choice to get out of a rut. However, if the photographer decides they want to try a new image-format (say, moving from 3:2 aspect-ratio dSLR to square photos using 6x6 medium-format film) then because what matters is the ability to visualise a given output and the best way to achieve that is to practice with a camera that enforces a square view of the world, it might make sense to learn to recognise possible square compositions by acquiring such a camera.

Kinds of Bracketing

Bracketing is the process of taking multiple photos at source with a view to choosing favourable results after processing. It can happen for several reasons:


This is the weakest - wasteful and careless - way to work. To take two simple cases, a photographer might take a handful of photos at varying exposures on the off-chance that one of them is about right; others will be deleted. Or again, they might take several photos of a closeup scene varying their scene/camera distance a few millimeters, but again, only choose one frame to go forward.

With data-processing in mind

This is the strongest methodology. If we have a tricky exposure - for example, very high contrast scene - we might choose to bracket 3 exposures around average plus or minus 1.5EV. The previous method would have us reject two of those and work from one. Of course, with the increasing use of HDR (High Dynamic Range) technology, we can now say that well-exposed pixels from all three source images contribute in some way to the finished output. Because more pixels have gone in, the results will be that much more accurate and the photographer has much greater flexibility in choosing exactly how the finished work should look.

Other techniques that can be deployed here are panorama stitching, stacking (for noise-reduction), focus-stacking (improves apparent depth of field) and super-resolution (uses sub-pixel alignment differences to upscale or interpolate inter-pixel values).

The data might come in handy

Not the strongest argument, but still valid: at the time of writing, the techniques of super-resolution and focus-stacking are fairly new to mainstream desktop image-processing. Who knows what delights are coming in the next 5 years?

Even if a finished image doesn't rely on new technology, if it's just a matter of available time to process a photo-shoot the same evening, it might be that another day will have enough time to process a panorama where initial processing only worked from one frame.

Results-bracketing / Stochastic Photography

It is a matter of pride that dictates that, as photographer, we like to have everything controlled, precisely designed from the first frame and reproducible and deterministic in infinitessimal detail. The phrase "get it right in-camera" and term "unmanipulated" are mantras in this field.

"Stochastic photography" is on an acknowledgement that some features of a scene are simply outwith our control.

For example, one might be making a photograph of a small water cascade flowing over and around rocks; we know that if a shutter-speed is chosen slow relative to the water flow-rate, then the water will come out resembling brushed white cotton wool.

If you make a handful of photographs of such a scene, you might find that split-second variations in rate and direction of flow lead the soft water highlights to accumulate obscuring some parts of a stone rather than others, or a rogue droplet might fly off leaving a trail through space; you can then choose a preferred version based on those criteria.

On the other hand, if you only make the one photograph at the scene, you'll not even know that those were considerations. It dramatically increases the risks of either making a sub-optimal photograph (potentially requiring a repeat visit to an out-of-the-way location) or even outright rejecting it for some reason.

It might be water/stone obscuration in the case of a tiny waterfall; it might be the degree of separation or interference of raindrop ripples in a puddle; it might be a gentle breeze amongst the trees disturbing the leaves to obscure a distant mountain; either way, the photographer is admitting something might change and is prepared to learn what features are important by choosing between alternatives later on.

(Note that this is not bracketing for technical error, only because some feature of the scene changes, so it includes an element of getting it right in-camera as well.)

blog comments powered by Disqus
light photography exposure art philosophy absolute relativism bending-the-rules methodology results creative shade