Every serious photographer has a folder — physical or digital, hidden or accessible — of images they do not show. Not images that failed technically, which can simply be deleted, but images that almost worked: that had the right subject, the right light, something of the quality they were after, and that nevertheless fall short of what they intended in a way they cannot always name. These images accumulate over years. They are among the most important things a photographer owns.
The instinct is to ignore them: to put energy into the images that succeed, to show those, to build a practice around strength rather than weakness. This is sensible as a strategy for presenting work to the world. It is a disaster as a strategy for developing the eye. The images that don’t work — specifically, the ones that almost work, that arrive at the edge of something and fail to cross it — contain more information about the current state of the photographer’s seeing than any number of successful images. They mark the boundary of what the eye can currently do. And the boundary is where development happens.
This chapter is about that boundary: about how to read the images that fail, how to distinguish between different kinds of failure, and how the patient, honest analysis of what went wrong in the nearly-right image produces, over time, the conditions under which the fully-right image becomes possible.
There are at least three distinct kinds of failure in a photograph, and they require different responses.
The first is technical failure: the image is not sharp, or not exposed correctly, or the color is wrong in a way that has nothing to do with the color the photographer saw. Technical failure is the least interesting kind, because its cause is usually identifiable and its remedy is usually learnable. It tells you something about equipment or process, not about seeing. Discard these without regret and move on.
The second is compositional failure: the photographer saw something worth seeing, and made the exposure at approximately the right moment, but the formal organization of the image does not hold. The weight is wrong. The planes are not in productive tension. The eye has no path through the frame, or too many competing paths, or a path that arrives nowhere. This kind of failure is far more useful than technical failure, because its cause is in the seeing rather than the equipment, and its analysis yields direct information about what the eye is not yet doing: what it is missing about visual weight, or layering, or compositional structure, that the successful image possesses.
The third kind — the most useful and the hardest to analyze — is imaginative failure: the image is technically accomplished, compositionally reasonable, and yet fundamentally inert. It is correct without being alive. It has organized the world without finding anything in it. This is the failure of a photographer whose technical and compositional skills have outrun their capacity for genuine attention: who can make a good photograph of almost anything but cannot yet make a necessary one. The analysis of imaginative failure does not yield a technical or compositional remedy. It yields a more uncomfortable question: what were you actually looking for when you made this image, and did you find it?
Technical failure tells you what you did wrong. Compositional failure tells you what you didn’t see. Imaginative failure tells you who you haven’t yet become.
Tarkovsky, in Sculpting in Time, writes about the necessity of suffering in the creative process — not as a romanticization of difficulty, but as an acknowledgment that resistance is information. The image or film sequence that does not work is not merely a failure to be overcome. It is a communication from the work itself about what it needs that the maker has not yet found. The resistance is real, and it is pointing at something: a formal problem that has not been solved, a quality of attention that has not yet been achieved, a truth about the subject that the current approach cannot reach.
This is an uncomfortable way to think about failure because it places the responsibility entirely with the maker. The image did not fail because of bad luck or difficult conditions or an uncooperative subject. It failed because the maker’s current capacities were not equal to what the subject required. That is a harder thing to accept than a technical explanation, but it is also, ultimately, a more useful one: it tells you not what to fix, but what to develop.
The useful image — the title of this chapter’s second subject — is a concept that requires some unpacking, because it runs against the usual way photographers evaluate their work. The usual evaluation is aesthetic: is this image good? Is it strong? Does it succeed? The useful image is not necessarily any of these things. It is the image that teaches you something about your current seeing — that marks, more precisely than any successful image could, the boundary of what you are currently capable of, and points clearly at what lies beyond it.
A useful image might be one that fails in a new way — that fails at something you have not previously attempted, which means the failure is evidence of development rather than stagnation. A useful image might be one that fails at what it was trying to do but accidentally succeeds at something else — revealing a formal possibility, a quality of light, a spatial relationship, that the photographer was not looking for and may not have noticed without the failure as a frame. A useful image might be one that was discarded years ago and that, returned to with the eye the photographer has since developed, reveals qualities invisible at the time.
This last category — the image that failed at the time of making and reveals its qualities only later — is worth pausing on. It is not uncommon. The photographer who returns to their archive after five or ten years of continued development frequently finds images they dismissed at the time that now appear, to the more developed eye, to have been doing something they were not yet equipped to recognize. The image did not change. The eye did. And the eye, looking back, can now see what it missed — which is itself a measure of how much it has developed.
Patience is the quality that underlies everything in this chapter and in this part of the book. Patience with failure. Patience with the slow, imperceptible development of the eye that means you cannot see, this year, what you will be able to see in five years. Patience with the practice itself — the long arc of making and looking and editing and failing and making again that constitutes a serious photographic life.
The eye develops slowly. This is not a problem to be solved by working harder or looking more or reading the right books — though all of these help, in their way. It is simply the nature of the development. The eye is changed by experience, and experience takes time. The photographer at forty sees things the photographer at thirty could not, not because they have learned more facts or mastered more techniques, but because they have spent ten more years in the presence of the world, attending to it with whatever quality of attention they could bring to bear at each stage, and the accumulated attention has changed the instrument itself.
This is the only kind of development that matters, in the end: not the acquisition of skills but the development of the eye. The skills are means. The eye is the end. And the eye, as this book has tried to argue throughout, is developed through sustained, honest, patient engagement with images — yours and others’ — and with the world that produces them.
The images that don’t work are part of that engagement. They are not embarrassments to be hidden or mistakes to be corrected. They are the record of where the eye currently is and what it is reaching for. Treat them accordingly: with the same patient, honest attention you bring to the images of the photographers you most admire. They deserve it. And so does the eye that made them.