A photograph of a pipe hangs in a museum. Beneath it, painted words declare: Ceci n'est pas une pipe—This is not a pipe. Magritte's famous paradox irritates and delights because it exposes something we rarely consider: the strange magic by which flat surfaces convince us we're seeing three-dimensional worlds.

We speak casually of pictures showing things, as if representation were simply a matter of holding up a mirror. But mirrors don't require interpretation. A child raised without exposure to photographs struggles to read them. Perspective drawing, that seemingly obvious technique for depicting depth, was developed in Renaissance Italy and remains foreign to many artistic traditions.

How do two-dimensional marks on canvas create the impression of space, volume, and presence? The answer reveals that pictorial representation is far stranger and more culturally constructed than our immediate experience suggests—and understanding this transforms how we see every image we encounter.

The intuitive theory of depiction seems obvious: pictures work because they resemble what they represent. A portrait looks like its subject. A landscape painting looks like the countryside. Simple enough—until you examine the claim closely.

Consider that a photograph of a sunset and the actual sunset share almost no physical properties. One is a thin layer of chemicals on paper; the other is electromagnetic radiation from a star 93 million miles away. The photograph is cold, flat, and fits in your hand. It resembles the sunset in precisely zero measurable ways. Yet we instantly recognize what it depicts.

More troubling for resemblance theory: anything resembles anything else in some respects. My coffee cup resembles the moon—both are round, both can cast shadows, both exist in three dimensions. Yet no one mistakes my cup for a lunar portrait. Resemblance is symmetrical and ubiquitous; representation is directional and specific. The portrait represents the sitter, but the sitter doesn't represent the portrait.

Artistic traditions worldwide demonstrate that radically different visual strategies can successfully depict the same subjects. Egyptian profile figures, Chinese scroll paintings, Aboriginal dot paintings, and Renaissance oil portraits all represent human beings—through utterly incompatible visual means. If depiction required copying appearances, only one tradition could be correct. Instead, we have countless successful representational systems, suggesting that something other than resemblance underwrites pictorial meaning.

Takeaway

When you find an artwork's style 'unrealistic,' pause before dismissing it—you may be mistaking your own learned conventions for universal truth about how pictures must work.

The philosopher Nelson Goodman proposed a radical alternative: pictures function like languages. Just as words bear no natural connection to their meanings—'dog' doesn't bark or wag—pictorial elements connect to their referents through convention, habit, and cultural agreement.

This explains why pictorial competence must be learned. Linear perspective, which strikes Western viewers as simply 'how things look,' required invention and instruction. When perspective images were first shown to people from cultures without this tradition, they often couldn't interpret them. The convergence of parallel lines, the systematic diminution of distant objects—these aren't natural perceptions but acquired skills for reading a particular symbolic system.

Consider how readily we accept impossible pictorial conventions. Cartoon characters display thought bubbles above their heads. Maps represent three-dimensional terrain through contour lines. Medical illustrations show transparent skin revealing underlying organs. None of these resemble actual visual experience, yet we read them effortlessly—because we've internalized their symbolic grammars.

The language analogy illuminates why artistic styles change historically and vary culturally. Just as languages evolve and differ, pictorial systems develop, transform, and coexist. Medieval manuscripts, Impressionist canvases, and digital interfaces all constitute coherent visual languages—each with distinct vocabularies, syntaxes, and expressive possibilities. Fluency in one doesn't automatically transfer to others.

Takeaway

Treat unfamiliar artistic traditions as foreign languages requiring patient study rather than failed attempts at your own visual idiom—the reward is access to genuinely different ways of seeing.

Once we recognize that pictures operate through conventions rather than natural resemblance, we can begin identifying the codes embedded in seemingly transparent images. Every representational choice—framing, angle, lighting, color—carries cultural freight that shapes meaning.

Take the convention of the gaze. In Western portraiture, subjects who meet our eyes assert presence and equality; those who look away suggest modesty, evasion, or availability for our scrutiny. These aren't universal meanings but specific cultural codes. Understanding them reveals how images construct relationships between viewers and subjects, often reinforcing social hierarchies invisible to casual glances.

Spatial conventions similarly encode ideology. The 'God's-eye view'—showing scenes from impossible elevated perspectives—historically associated such images with divine or authoritative knowledge. Low angles aggrandize subjects; high angles diminish them. These aren't natural effects but learned associations, deployed (consciously or not) to shape our responses.

The seemingly neutral conventions of photographic realism prove especially important to decode, precisely because they appear unconventional. Photographs seem to show 'what was there,' masking the countless choices—film stock, lighting, cropping, timing—that construct their meanings. Recognizing photography as another symbolic system, no more inherently truthful than painting, inoculates against the manipulation that exploits our trust in photographic transparency.

Takeaway

Before any image, ask: what conventions is this using, and whose interests do they serve? The most powerful visual codes are those that have become so familiar they seem like simple reality.

Magritte was right: the pipe was never a pipe. Every picture operates through symbolic conventions that transform marks into meaning—conventions as arbitrary and culturally specific as any spoken language, however natural they feel to fluent readers.

This recognition doesn't diminish pictorial power but deepens our appreciation of it. Understanding that perspective is an invention rather than a discovery, that photographic realism is a style rather than a transcription, enriches rather than impoverishes visual experience.

The flat surface doesn't deceive us; we collaborate with it, bringing our learned competencies to the interpretive encounter. Every image becomes an invitation to recognize both what we see and how we've learned to see it.