Bad Art Advice

The internet may be the greatest art classroom ever assembled. Within seconds, an aspiring artist can watch someone paint a portrait, prepare a panel, mix a flesh tone, sharpen charcoal, construct a perspective grid, arrange a still life, varnish a painting, or explain why a composition supposedly "works." Techniques that once required access to a particular teacher, school, studio, or book are now available almost instantly. That's extraordinary. It also creates a problem.

The same systems that allow good information to travel farther and faster allow questionable information to do exactly the same thing. A confident explanation can be repeated by thousands of artists until it begins to sound like established knowledge. A preference becomes a recommendation. A recommendation becomes a rule. A rule acquires an explanation. Eventually, few people remember to ask whether the explanation was ever demonstrated in the first place.

Art education is particularly vulnerable to this, not because artists are unusually gullible, but because art sits at an unusual intersection of subjective preference, physical materials, human perception, learned skill, cultural convention, and creative intention. Statements belonging to one of these categories are routinely mistaken for statements belonging to another. "I prefer not to center my subjects" is very different from "You should never center your subject." "I find this palette useful" is different from "You need this palette to paint realistic color." "This procedure works well for me" is different from "This is how oil painting works." And "try this" is very different from "this is true."

That difference is where a great deal of bad art advice begins.

What Do We Mean by "Bad"?

Calling something bad art advice can easily become an exercise in taste. One painter loves glazing. Another prefers direct painting. One artist works carefully from observation. Another constructs images almost entirely from imagination. Someone paints with three colors; someone else uses twenty. Some artists love elaborate preparatory drawings while others begin immediately on the painting surface. None of these differences establishes that one artist possesses the correct way to make art. Creative goals vary enormously, and so do the procedures artists develop to pursue them.

So the problem I want to address is not, "Are you making art the way I think you should?" The more useful question is, "What exactly are you claiming to know?"

That question changes everything. An artistic preference can be perfectly legitimate without being universally preferable. A heuristic can be extremely useful without being a law. A convention can facilitate communication without being biologically necessary. A particular procedure can work without the explanation accompanying it being correct. And a physical property of a material does not disappear because an artist dislikes being told what to do.

Consider the difference between saying, "Try placing the subject off-center," "I usually prefer subjects placed off-center," "Subjects should be placed off-center," and "Off-center subjects are inherently more interesting." These statements may sound similar, but they make progressively stronger claims. The first proposes an experiment. The second reports a preference. The third prescribes behavior. The fourth makes a general claim about aesthetic response. Only the latter claims demand substantial justification. Unfortunately, online art education frequently slides from the first statement to the last without anyone noticing.

The Great Transformation: From Preference to Law

Artists understandably search for repeatable solutions. Representational image-making presents an enormous number of interacting variables. Shape, value, color, edge, orientation, scale, position, texture, depth, lighting, materials, viewing conditions, and countless other factors may contribute to what we eventually experience as an image. Faced with that complexity, simple rules are attractive.

Put the focal point here. Never divide the canvas in half. Avoid tangents. Use odd numbers of objects. Warm colors advance. Cool colors recede. Never use black. Always simplify. Never blend. Every edge needs variety. The eye enters here. The eye exits there.

Some of these statements may occasionally function as useful prompts. Some may describe tendencies under particular conditions. Some may simply reflect an instructor's preferred way of working. The trouble begins when their conditions disappear. "Try this under these circumstances" becomes "do this." Then "do this" becomes "good artists do this." Finally, someone invents an explanation for why everyone must do it.

At that point, a heuristic has quietly transformed into a law.

The Rule of Thirds Problem

Composition provides an almost ideal demonstration. The Rule of Thirds is probably among the best-known compositional prescriptions in contemporary art and photography. Divide the picture into thirds horizontally and vertically, we are told, and important elements should be placed along those divisions or near their intersections. There is nothing inherently wrong with doing this, nor is there anything wrong with using a grid to encourage yourself to explore placements you might otherwise overlook. The problem appears when the grid is assigned powers that have not been established.

The historical "Rule of Thirds" associated with John Thomas Smith's Remarks on Rural Scenery from 1797 was not originally identical to the familiar modern nine-part-grid prescription. Smith discussed unequal divisions of light, shade, land, sky, line, form, and color, and argued for roughly one-third/two-thirds relationships over equal divisions. Later teaching transformed the idea into something much more rigid: important subjects belong near the thirds, while centered placements are often characterized as static, boring, amateurish, or aesthetically inferior.

Now we have something that can actually be investigated. If thirds locations possess a general aesthetic advantage and centered positions a general disadvantage, people should exhibit preferences consistent with those predictions under appropriate experimental conditions. Direct experimental investigations of object placement have not supported the strongest version of that claim. Research examining placement preferences instead indicates that judgments can be influenced by factors such as object direction, position, perspective, and the structure of the frame. Direct experimental testing of Rule-of-Thirds predictions has also produced results inconsistent with its strongest claims.

This doesn't mean nobody should use thirds. It means something much more important: a compositional device can be useful without its mythology being true.

Perhaps placing something near a third works wonderfully in your picture. Fine. But the fact that it works does not demonstrate that it works because it occupies a third. That distinction is enormous.

Finding the Pattern After the Painting Exists

One reason compositional mythology survives is that completed artwork provides endless opportunities for pattern hunting. Take a successful painting and overlay enough lines on it, and eventually something interesting will happen. A diagonal intersects an eye, a spiral curls toward a hand, a rectangle surrounds the important objects, a triangle connects three heads, or a golden section passes through something significant. The result can look astonishingly persuasive.

But finding a pattern after the fact is not the same thing as demonstrating that the pattern caused the result. If we examine enough successful paintings, we can undoubtedly find many containing thirds, diagonals, triangles, spirals, rectangles, radial structures, symmetry, asymmetry, centered subjects, off-center subjects, repeating intervals, and countless numerical relationships. Successful paintings also contain pigments. That doesn't mean pigment causes aesthetic success.

Correlation alone doesn't establish causation. The more flexible the compositional system, the easier retrospective confirmation becomes. If we can move, rotate, resize, subdivide, extend, or select among numerous geometrical structures until something aligns with something else, a match becomes increasingly unsurprising.

The useful question isn't, "Can I find this pattern in successful art?" It is, "Does deliberately applying this pattern reliably produce the claimed advantage compared with relevant alternatives?" Those are very different questions.

But Composition Isn't Arbitrary Either

Rejecting imaginary rules does not require us to conclude that composition is meaningless or completely subjective. That would simply replace one bad explanation with another.

Pictures are visual stimuli, and visual systems have limits and tendencies. Contrast matters. Scale matters. Visibility matters. Spatial relationships matter. Orientation matters. Context matters. Attention matters. Previous experience matters. If an intended element is too small to resolve under the relevant viewing conditions, information can be lost. If two areas provide insufficient contrast for discrimination, their intended distinction may become difficult or impossible to perceive. If visual relationships produce unintended grouping or ambiguity, communication may change.

These aren't commandments about where to put a tree. They are constraints and relationships associated with visual communication. That suggests a much richer conception of composition than arranging objects according to a universal template.

Composition can instead be approached as a problem of information management: what information is present, where it appears, how elements relate, what viewers can discriminate, what attracts or sustains attention, what expectations are engaged, and how those variables interact with the artist's intended communication. Instead of asking, "Where does the rule say my focal point should go?" we can ask, "What am I trying to communicate, and what happens when I put it here?"

Now composition becomes an investigation rather than a ritual.

Materials Don't Care About Your Philosophy

Things become even more interesting when we move from composition into materials. Here, "there are no rules in art" can become genuinely misleading. There may be no universal aesthetic law requiring you to paint thinly, thickly, smoothly, loosely, transparently, opaquely, brightly, or darkly. But materials have properties. Oil paint doesn't negotiate those properties according to artistic ideology.

Drying oils undergo chemical changes as they cure. Pigments differ in characteristics such as particle structure and oil demand. Grounds differ in absorbency. Layers can differ in flexibility. Solvents affect paint handling and, depending on their use, can influence binder concentration. Paint films can wrinkle, crack, delaminate, become underbound, sink, yellow, or otherwise change with time. Those aren't stylistic opinions.

This creates an important distinction between statements such as "You should never paint this way because it looks bad" and "This material combination may increase the probability of this physical outcome." The first is largely aesthetic; the second is a technical claim that can potentially be investigated through chemistry, materials science, conservation evidence, and observation of aging paint films.

Consider the traditional oil-painting guideline commonly summarized as fat over lean. Whatever oversimplifications may accompany the phrase in casual teaching, it concerns relationships among successive paint layers, oil content, drying behavior, flexibility, and long-term paint-film behavior. That is categorically different from saying, "Never put your focal point in the center."

Calling both statements "art rules" obscures an important difference. One is attempting to describe material behavior; the other is prescribing a compositional choice. Breaking the latter may produce an image you love. Ignoring the former may produce a paint film that behaves differently than you intended.

Creative freedom does not require pretending chemistry is optional.

The Recipe Problem

Another common source of bad advice is the recipe. Recipes are seductive because they

Recipes can be useful. In fact, for a learner confronting too many variables at once, reducing the number of choices may make a problem considerably more manageable. But successful execution of a recipe is not necessarily understanding.

Suppose an artist follows a tutorial and produces an attractive portrait. What has been demonstrated? At minimum, the artist demonstrated an ability to follow that procedure under those circumstances. But can the artist diagnose what went wrong when the lighting changes? Can the procedure be adapted to another complexion, surface, environment, or illumination? Can the artist explain which portions of the procedure are necessary, which are optional, and which merely reflect the demonstrator's preferences? Can the artist achieve a comparable outcome through another procedure?

A tutorial may teach someone how to reproduce a sequence while leaving the underlying problem largely untouched. This distinction matters enormously for representational artists. The goal cannot simply be to accumulate more recipes for situations already encountered. Useful skill must increasingly support problem-solving when the exact answer has not already been demonstrated.

Watching Is Not the Same as Learning

Online art culture can also create an illusion of practice. Watching a skilled painter work is fascinating. It can reveal procedures, tool handling, sequencing, decision-making, and possibilities that a learner may never have considered. But watching someone solve a problem is not equivalent to developing the ability to solve that problem yourself.

Skill development requires doing, and mere repetition isn't necessarily sufficient either. Research on skilled performance distinguishes casual or naive repetition from more structured forms of practice. Productive training involves defined goals, challenges appropriate to the learner's current capabilities, informative feedback, attention to errors, and repeated attempts modified in response to what happened previously.

That distinction should make artists suspicious of another familiar piece of internet advice: "Just draw every day." Drawing every day may be wonderful. It can build familiarity, maintain established capabilities, produce artwork, encourage exploration, or simply provide pleasure. But frequency alone tells us very little about what is being learned. An artist can repeat the same ineffective strategy every day, and repetition can reinforce habits as readily as it can modify them.

A more useful question is not simply, "How much are you practicing?" but "What is changing as a result of your practice, and what feedback is informing that change?"

"Draw What You See"

Few pieces of representational art advice sound more reasonable than "Draw what you see." There is useful intent behind the phrase. It often encourages students to attend to available visual information rather than substituting simplified conceptual symbols, drawing the generic "eye" they think they know instead of examining the particular shapes and relationships present in front of them.

Taken literally, however, the statement hides an enormous problem: what exactly do we "see"?

Human vision isn't a neutral measuring instrument delivering a direct copy of the physical world. Perceived brightness does not map simply onto measured luminance. Perceived color depends upon surrounding conditions and adaptation. Apparent shape and size interact with depth, orientation, context, and interpretation. Attention is selective. Visual information is incomplete. The retinal stimulus itself does not uniquely specify every property of its physical source.

Seeing is already an active biological process.

For representational artists, this is not philosophical trivia. It helps explain why apparently simple observational tasks can be surprisingly difficult. The artist isn't copying an objective picture conveniently projected into consciousness. The artist is attempting to construct a physical surrogate capable of producing useful perceptual relationships under another set of viewing conditions.

That is a considerably more interesting problem than "paint what you see," and it helps explain why representational skill involves learning to make increasingly useful discriminations, comparisons, predictions, and corrections, not merely trying harder to look.

Beware of Neuroscience-Flavored Art Advice

The internet has discovered that almost any art claim sounds more authoritative when the brain is involved. "The brain prefers this." "The eye naturally goes here." "This activates the right brain." "Humans are hardwired to find this proportion beautiful." "Warm colors come forward because the brain reads them as closer." "The eye enters the painting from the left."

Statements like these should immediately provoke a simple response: How do we know?

Invoking the brain is not evidence. Everything we visually experience necessarily involves the nervous system in some fashion. Merely attaching neurological vocabulary to an artistic preference explains almost nothing.

A meaningful scientific explanation requires more. What phenomenon was measured? Under what conditions? Compared with what? How large was the effect? Was the finding replicated? Does the study actually test paintings, or are conclusions about paintings being extrapolated from a much narrower experimental task? Does the evidence demonstrate a general tendency, or is someone converting a conditional effect into a universal prescription?

These questions aren't hostile to art. They're hostile to pretending.

The Algorithm Rewards Certainty

There is another reason bad art advice flourishes online: nuance is difficult to package. "NEVER DO THIS IN YOUR PAINTINGS" makes a better thumbnail than "Under certain conditions this strategy may produce an undesirable result depending upon your intention." "THE SECRET TO PERFECT COMPOSITION" travels farther than "Here are several interacting variables that may influence judgments of spatial arrangement."

Social media rewards compression, novelty, confidence, immediacy, and repeatability. Knowledge does not always possess those characteristics. Sometimes the most responsible answer is "It depends." Sometimes it is "We don't know." Sometimes it is "This appears useful, but the explanation usually given for it is questionable." And occasionally it is "There is evidence for this effect, but it is much narrower than people online are claiming."

Those statements don't fit comfortably on a thumbnail. But intellectual responsibility frequently begins where the thumbnail ends.

The Demonstration Trap

There is also a subtle error that occurs when accomplished artists teach. A skilled artist demonstrates a procedure and produces a beautiful result. The result itself then seems to validate everything the artist said while producing it.

But artistic performance and explanatory knowledge are not identical.

Someone can be extremely capable at performing a task while misunderstanding some of the mechanisms responsible for the outcome. A painter may possess years of highly developed perceptual and motor experience. They may make extraordinarily effective decisions rapidly and intuitively. But when asked why a particular decision works, the explanation may be reconstructed afterward from tradition, personal theory, studio folklore, or whatever explanation they were originally taught.

The painting can be excellent while the explanation is poor. This is not an insult to the painter. It is simply a reminder that being able to do something and being able to correctly explain why it happens are different capabilities.

Great artwork is evidence of great artwork. It is not automatic evidence for every theory held by the artist who made it.

So What Should Artists Trust?

There is no single authority capable of screening every piece of art advice for us, but we can improve the questions we ask. When encountering an instructional claim, first determine what kind of claim it is. Is it a preference, a heuristic, a convention, a pedagogical strategy, a claim about perception, a claim about materials, a historical claim, or a claim about human behavior? Those categories demand different kinds of support.

Then ask what would happen if the advice were violated. If someone says, "Never center a portrait," what exactly is predicted to happen? If someone says, "Never use black," what measurable or observable failure is supposed to follow? If someone claims that a golden spiral produces beauty, what evidence distinguishes that explanation from the thousands of other geometrical relationships that could be retrospectively fitted to the same image? If someone recommends a painting medium, what does each ingredient actually do? If someone says a technique is "traditional," does that establish historical usage, physical stability, aesthetic superiority, or merely that people have used it before?

And perhaps most importantly: What evidence could change the person's mind? If the answer is "nothing," we probably aren't dealing with a testable explanation.

Good Advice Can Be Modest

Interestingly, good art advice often sounds less impressive. "Try this." "Compare these." "Watch what happens when…" "This works reliably for me under these conditions." "This material behaves differently when…" "This is a convention rather than a physical requirement." "We use this exercise to isolate this particular problem." "This explanation is uncertain." "I don't know."

Those phrases leave room for investigation. They also give the learner something far more valuable than obedience: a way to evaluate outcomes.

The purpose of good instruction should not be to fill an artist's head with commandments. It should increase the artist's capacity to make informed decisions.

Demystification Is Not the Enemy of Creativity

Whenever discussions like this arise, someone understandably worries that analyzing art too closely will somehow drain the magic from it. I think the opposite is true.

Understanding pigment doesn't prevent someone from being moved by color. Understanding visual perception doesn't make seeing less extraordinary. Understanding skill acquisition doesn't eliminate intuition. Understanding compositional relationships doesn't prevent surprise. And discovering that a favorite artistic "rule" has little evidential support doesn't take anything away from the paintings made with it.

Artists remain free to use thirds, use spirals, center everything, center nothing, paint with black, never touch black, use three pigments, use thirty, glaze, scumble, paint alla prima, measure everything, eyeball everything, or break every convention available. Creative freedom means those choices remain available.

But freedom becomes more useful when we understand what kind of choices they are.

The danger isn't experimentation; it's confusing experimentation with evidence. It isn't tradition; it's confusing tradition with necessity. It isn't preference; it's disguising preference as principle. It isn't a heuristic; it's forgetting that the heuristic was only a heuristic. And it certainly isn't creativity; it's replacing curiosity with commandments.

A Better Question

The next time someone tells you that a painting works because the focal point sits on a third, that you must never use a certain pigment, that the eye always travels through a composition in a particular direction, that a mathematical ratio produces beauty, or that one simple trick will transform your drawing, you don't necessarily need to reject the advice.

Just ask a better question: What exactly is being claimed, and how would we know?

Those two questions won't tell you how to make art, and thankfully so. But they may help you distinguish between a preference, a useful tool, a physical constraint, a teaching strategy, a testable principle, and a very confident story.

In an age with more art instruction available than at any other point in history, learning to make those distinctions may be one of the most useful skills an artist can develop. produce clear instructions: add this much medium, use these three colors, start every portrait with this mixture, glaze with this, use this brush, paint the shadow this color, make the background warmer, blur this edge.