The Art and Science of Photography: The Rule of Thirds Is Dead. Long Live the Rule of Thirds.

Has the rule of thirds influenced your compositions? It has mine at times. Yet, with each application I wonder, why should placing things near the thirds of an image be appealing? In fact, it turns out that the link between the rule of thirds and aesthetic appeal is weak. Is the rule of thirds dead?

A Little History

This most common of compositional techniques was initially proposed by John Thomas Smith, Keeper of Prints at the British Museum. He published a book in 1797 entitled “Remarks On Rural Scenery” within which he laid out the notion of a rule of thirds:

In short, in applying this invention, generally speaking, to any other case of light, shade, form or color, I have found the ratio of about two thirds to one third, or of one to two a much better and more harmonizing proportion, than the precise formal half, the too-far-extending four-fifths — and, in short, than any other proportion whatever.

(On a side note, do you think Mr. John Thomas Smith would be a bit surprised to find that, two hundred and some odd years later, half the population of the planet is carrying around a little device in their pocket called a “phone” that helpfully has his rule of thirds grid overlaid right on its “camera”? Oh, the missed royalties… but I digress.)

The question that keeps coming to my mind, though, is why should there be a “harmonizing proportion”, a rule of thirds? Why should we innately find it appealing to place prominent subjects or boundaries a third of the way into an image? Smith’s hypothesis was that “parts of equal appearance hold [the attention] awkwardly suspended as if unable to determine which of those parts is to be considered as the subordinate.” It’s actually an interesting proposition. Many of Ramachandran’s neuroaesthetic laws, for example, are motivated by the need for our brains to effectively divvy up our very limited attention.

Yet, it seems like one can come up with contrary examples, examples that would suggest Smith’s explanation might not be the whole story. What if we had an aesthetically strong image that was composed according to the rule of thirds, yet, within which the rule of thirds wasn’t able to play a role in apportioning our attention? It might suggest that there were other factors at play beyond Smith’s explanation (or that the rule of thirds was irrelevant).

The rule of thirds doesn’t help us to apportion our attention in this case, as two equally strong subjects lie on the image’s thirds. To my eye, this doesn’t lessen the strength of the image in the slightest. This would suggest that Smith’s explanation for the rule of thirds may not tell the whole story. Image by Mark Dunsmuir | www.markandnaomi.com | instagram.com/mark_and_naomi.

Take a look at the image above, for example. It’s a crop from a larger portrait taken by fellow Staff Writer, Mark Dunsmuir. The image has two competing subjects, each of identical visual weight and each placed identically with respect to the rule of thirds. The rule of thirds, therefore, can suggest nothing about how we should divide our attention between these two focal points, yet, it still seems to contribute to a strong image (along with many other factors). I, thus, suspect there may be something to the rule of thirds, but am a little skeptical of Smith’s explanation.

In fact, nearly since its inception, there’s been a good bit of pushback against the rule of thirds, pushback that seems to go beyond your typical artist’s nonconformist streak. That wariness turns out to be well founded. In a recent study integrating the computational analysis of images with the human perception of aesthetic beauty, researchers found that the rule of thirds was only weakly correlated with the aesthetic appeal of an image. And, perhaps tellingly, I haven’t yet been able to find a good neuroaesthetic motivation for the rule of thirds in the literature. Yet when composing a shot, it often doesn’t seem a bad place to start, either. Rule of thirds-based compositions do seem to be more appealing than many other options. Perhaps the rule isn’t sufficient on its own to yield superbly crafted compositions, but maybe it’s enough to help save us from some much worse ones. Indeed, in the study referenced above the least aesthetically appealing images all had low rule of thirds scores. There did appear to be a correlation there.

I’ve, thus, spent a good bit of time over the last few weeks wondering what in our evolutionary history might have yielded a preference for image compositions based on thirds. My initial thought was that maybe this sort of framing was somehow ideal for extracting three-dimensional information from a scene in the real world; or maybe such compositions meet the competing goals of allowing us (the viewer) to keep an eye on a subject (say, the person we’re talking to) while still directing the main focus of our attention to the distant background (say, in the direction a lion might be approaching from). My wife quickly dismissed both as a little hare-brained; and they do both have tragic flaws. My other ideas fizzled out, too.

But it did get me thinking. I eventually started to wonder if maybe we’re asking the wrong question. Let’s take a quick look at a couple of other popular “rules”.

Don’t Put the Subject in the Center of the Frame

There’s another rule of thumb we often hear in photography: don’t put the subject in the center of the frame. It turns out that this is good advice. In most cases, people tend to prefer that objects not lie on the centerline of an image. There is, however, an exception. We do prefer forward-facing, symmetrical objects to be centered (see, for example, the image below).

One of the few cases where a centered subject can be pleasing: when the subject, itself, is symmetrical and forward-facing.

Why should this be the case, though? Why should we prefer that symmetrical objects be centered and asymmetrical objects be moved off-center? If I were to hazard a guess, it would be that placing something on the centerline, the frame’s axis of symmetry, invites our brains to investigate just how mirror-like that symmetry actually is. And if it’s not, we come away disappointed. The image above works fairly well precisely because the subject is symmetrical. We’re critically tuned to recognize and respond to that symmetry. It’s deeply ingrained in our evaluation of beauty in human faces and, I suspect, in those of baby boobies as well. Evolutionarily, it’s thought to be a key indicator of good health. (In fact, the breaking of symmetry readily explains the plague that the bacterium, Propionibacterium acnes, represents for the teen dating scene. It makes faces asymmetric which alerts potential suitors, correctly, to an underlying health issue.)

This tendency of the centerline to highlight image symmetries can be used to great effect when we want to emphasize those symmetries in our images, as in the case of the juvenile Nazca Booby, above, or of the driftwood and building thunderhead, below.

In this image, the placement of the base of the driftwood in the center of the frame deliberately invites a comparison with the symmetrical shape of the thunderhead.

If the primary subject of an image isn’t symmetrical, however, and it's placed on the centerline, it's likely to invite a comparison that doesn't work in our favor. Check out the pair of images below, which differ only in the way they’re cropped.

The image on the left invites the brain to look for symmetry that’s not there. My eye moves first to the primary subject of the image only to be disappointed by the lack of symmetry. It then moves to the image background searching for symmetry there, but there’s none to be found. The mountains on the left are both darker and fill more of the frame than those on the right. The composition highlights the image’s asymmetry, its imbalance. It leaves me feeling unsettled. On the other hand, placing the subject a bit off-center to the left has a number of advantages. First, our brains no longer have any expectation of symmetry in the primary subject (which would only disappoint them). Second, the darker subject on the left and the brighter open sky on the right become competing centers of visual mass that are, themselves, symmetrically balanced across the centerline of the image. Finally, it gives the subject, who’s facing to the right and obviously in the process of walking, a space to “walk into”. Research has found, in general, that when an object is off-center and has a potential directionality, we tend to prefer that the object “face” toward the center of the frame.

For most subjects there is, thus, an aesthetic preference for off-center placement. Aesthetically appealing placements along the centerline are the exception.

Image Framing

Let’s think about one other aspect of image composition. The notion of an image frame, or boundary, has been implicit throughout the entirety of this discussion. But why should we have any innate, aesthetic response at all to how something is framed? We didn’t evolve to look at pictures. Throughout the vast majority of our evolutionary history there weren't paintings or photographs to think about the framing of. Why should we have any response at all to where the bounds of an image fall relative to the subjects within it?

Just because paintings and photographs haven’t been ubiquitous throughout most of our history, doesn’t mean our visual scenes haven’t been framed. Imagine sitting in a shelter on the African savannah waiting for a meal to cook when out of the corner of your eye you catch sight of a hyena moving in the tall grass; a hyena which, from your vantage point, is just at the edge of the door frame about to disappear from view behind the shelter wall. Would you think to yourself, “Ohhhh, isn’t that lovely? That’s just where I like to find large predators: nearby and about to disappear from my line of sight.”

A hyena walks out of the visual frame.

Of course, not. You’d be on your feet in an instant trying to gain a better visual, gain more context. Where is the hyena coming from? Why is it going in that direction? Is it alone? Are there other hyenas nearby, trailing it? Or are there other hyenas in front of it that might already be circling around on you?

We have a deep, emotional, survival-based need to have enough visual space to really understand the critical elements of our environment, their surrounding contexts, and the places they may be headed to next. Our lives may depend on it. So it doesn’t matter what’s framing your field of vision, whether it’s a shelter wall, the mouth of a cave, the open space between adjacent trunks within of a copse of trees, or the bounds of an image. Your likely to feel more emotionally at peace if there's a little distance between the hyena and the point at which it disappears from view.

What Do Symmetry and Framing Have to Do with the Rule of Thirds?

Let’s step back for a moment and see where this leaves us. We’ve discussed two different rules that roughly reduce to: don’t place subjects on the centerlines of the frame; and don’t place subjects too close to the edges. The “exclusion” zones suggested by the two rules are roughly indicated in the figure below.

Regions of the frame within which to avoid placement of subject(s) are indicated in red. Avoidance of image centerlines is shown at top-left. Avoidance of frame boundaries at top-right. The combined “exclusion” areas are shown at bottom-left. A comparison with rule of thirds guides is included at bottom right.

Note that when the exclusion areas of the two rules are combined, we end up with something that looks suspiciously like the rule of thirds. What if the rule of thirds isn’t an affirmative rule, in the sense that there’s some evolutionary motivation for why we should prefer the focal points of an image to lie along its thirds? What if it’s just shorthand for a couple of other negative rules, rules which do have evolutionary, neuroaesthetic roots?

This might explain some of the insensitivity we often experience with respect to a precise alignment with the rule of thirds. The notional exclusion zones illustrated above suggest a good bit more room to maneuver. It’s not so much about precise placement on a third that’s important, but just a fuzzier admonition not to get “too close” to the important boundaries or planes of an image, whatever “too close” means.

It might also explain, at least in part, why the correlation between the rule of thirds and human assessed aesthetic appeal mentioned earlier is weak. If we understand the underlying rules, and the reasons for them, there are myriad ways in which the rules can selectively be pushed or broken to achieve a particular end. An experienced photographer will likely be adept at using this greater compositional flexibility to strong aesthetic effect, moving beyond simple rule of thirds compositions.

Yet, it's still not a bad rule of thumb, especially for the budding photographer. It can help avoid many significant compositional pitfalls. Eventually, however, it's probably worth understanding something about the more basic considerations that may underlie image composition so that as we grow as photographers we can choose how and when to effectively break it. Long live the rule of thirds!

Thoughts? Other ideas? Let me know in the comments, please!

Images used with permission where applicable.