Perception is a curious phenomenon. Our mind continuously extracts patterns from our sensory input and processes them in strange ways.
Let’s take an example: we have a stroll through a tropical forest and suddenly we sense a movement. Our brain immediately starts working: ‘Teeth’, ‘Stripes’, ‘Tiger’, ‘ Run!!!’. Besides the identification of a large cat (tiger), we might also identify aggression (bared fangs, movement towards us).
It seems obvious from my example, how evolutionary useful this pattern recognition is and how it thus might evolve through natural selection. We can also assume, that many animals as part of their instinctive behaviour might posses the ability to recognise such patterns.
Philosophy has forever struggled with the phenomenon. One especially puzzling question was how our thinking about objects (patterns) is related to reality. For Plato, there were ideal forms (like the idea of a tiger) that we recognised in an concrete entity, that is the concrete animal. Aristotle thought that real things belong to categories of being that are reflected in our mind. Kant finally told us that such categories are just ways how our mind works and we don’t know anything about the real objects beyond that.
There is of course a first question, whether the concept of a tiger exists in our mind even without language or whether mental concepts and language are linked. If we consider that our thinking mostly happens through an inner monologue and our earliest memories typically date back to the time we acquired language, some link seems likely.
Experiments have also shown that mental associations differ, depending on the language, in particular the genus of a noun in the mother language of a test subject: Germans and Spaniards tend to associate different adjectives with nouns that have a different genus in these languages (the word for bridge, e.g. is feminine in German and masculine in Spanish). That of course leaves the question open for e.g. English or Finnish speakers. My point is: language seems to at least influence how we think about objects and the world.
Clearly, however, also visual recognition plays a role besides language. When in the German Bauhaus archetypal forms for use in universally recognisable signs were investigated, there were indicators that certain visual elements were necessary, e.g. to evoke the picture of a tree. Even here, though, there might be cultural differences and the archetype of a tree in Central Europe might more resemble an oak with thick stem and broad crown, while in the Northern Countries the tree archetype might have a more slender form. Still, if a shape contains a stem and branches, and maybe some leaves, it might be universally recognised as a tree.
This pattern recognition is, however, notoriously error prone because it seems optimised for speed (see the tiger example why this might be useful) rather than correctness. Daniel Kahneman’s book Thinking, Fast and Slow explains this in greater detail. But there are other pitfalls. One of them is expressed in the Sorites Paradox. This goes back to an obscure Greek philosopher (sorites is the word for “heap” in Old Greek) and asks: “If one has a heap of sand, how many grains of sand can be removed before the heap ceases to be a heap?”
If one removes one grain from a heap of a million grains of sand, the rest certainly still qualifies as a heap. But if all the grains except one are removed; does this single grain still qualify as heap? Probably not; but where goes the borderline?
There have been an astounding amount of solutions proposed to this problem (Wikipedia is your friend, if you are interested). Lawyers, of course, solve this through a normative approach: one second before the 18th birthday one is underage (in certain countries) and one second later empowered with full legal capacity, but a person hardly matures in that second. Such line drawing is eminently useful but doesn’t tell us about the underlying reality.
Languages (and maybe even our thinking) tend to treat certain patterns as distinct even if they aren’t (like the heap). Let’s look at another example:
Above, you see the visible electromagnetic spectrum of light, that means the wavelengths of light our eyes can see. Where exactly would you place blue, green, yellow, red in that picture?
In artificial intelligence, the problem has been approached by so-called fuzzy logic that leaves the world of distinct terms and introduces gradations. A pile of sand grains thus would increase in “heapishness” and our spectrum would have gradations and areas of “blueness” and so on.
Still the fundamental problem remains: the very mechanisms evolved to secure survival hinder our correct understanding of the world. This has significant practical implications. Certain diseases might be better treated if we look at our bodies not as entities but as ecosystems of microbiomes. The more we learn about nature, the more distinctions disappear and are replaced by gradations and interconnectedness. Where e.g. do the “hosts” end and the “symbiotes” begin or are they better understood as ecosystems as well? Every single cell in our body (or in a plant) goes back to three different bacterial ancestors that partially still reproduce independently. A significant part of our body weight is made up from the symbiotic and opportunistic microbiome in our gut. Are we rather lumps of bacteria or “distinct” entities?
Not for the first time regarding this blog you might have asked yourself what this has to do with photography. Well, photography has prided itself with the degree of “realism” it provides. But all it seems to do is to trigger pattern matching mechanisms. If I make a photo of a tree, all a viewer might notice is that “a tree is a tree, is a tree, is a tree”. But a tree, I think, cannot be understood without the winds shaping it, the precipitation and light enabling or hindering its growth, its symbionts like the mycelium or the other animals and plants it interacts with. In portraits of persons, photographers try to capture hints to the inner life of a person; is this possible for a tree?
And how in a photograph can gradations be captured, be shown that a landscape morphs from one state to another, is like a living thing? The photographer, I think, has to constantly stay aware that s(he) is struggling against the strong forces of pattern recognition and categorisation at work in our minds. Deeper insights will always, it seems, require the photographer to bend the perception of the viewer, to make the seemingly certain look uncertain. Oh my.