One or two subjects?

[Manfred M]

A few of us had the beginnings of a short discussion on what a single subject is. Does each main element in an image act as a single subject or do they become a single subject when the objects are in close proximity or when they touch?

The discussion started in this thread: Just add smoke + vases


Let me try to demonstrate my point of view, where objects that are close together or touching visually become a single subject.



Image 1: Objects physically touch / overlap





Image 2: Objects are physically close together, but are not in direct contact





Image 3: Objects are physically distant, but are still (somewhat) connected to each other by the grass





Image 4: Objects are physically more distant and are far enough apart so that the grass no longer visually connects them





Image 5: Remove the grass and there is no question that the two objects are visually distinct.




In my view, everyone will have a different view to when the main objects become two differnt subjects.

[DanK]

In my view, everyone will have a different view to when the main objects become two differnt subjects.

That was in part what I meant. This is continuous, not binary, and I don't think it's all that helpful to make the distinction.

Is the photo below of one subject, two (treating the two closer together as one), or three? I don't think it's a helpful question. The useful question is whether the larger distance between the two on the right compared to the two on the left--the larger gap on the right--is a plus or a minus. (The few people who ventured an opinion didn't agree anyway.)

[Manfred M]

Dan - First of all, there will always be disagreements on at what point this is an issue. One of the most common issues we see with many images is that they lack an main subject, area of interest or "landing zone".

The other one we often see is two (or more) main subjects competing for viewer attention. That is also something that tends to not result in the strongest composition, again something that we see often in mages. That was the point of this series of images.

Your image is similar to the one where this discussion started. You have a cluster of subjects that are in close proximity that visually present as a single main subject. In my case, there was a cluster of two subjects that present as a single main subject. That is why the images work. Had the subjects been farther apart, we both would have introduced separate main subjects that would not have been as effective. My argument in that image was simply that you have look at an image based on how the subject is constructed (rule of odds doesn't necessarily hold).

A good example that I use in some of my talks is this 1926 image by August Sander (who is often credited as the first deadpan portrait photographer) showing "The Architect Hans Heinz Lüttgen and his Wife Dora". Two people, but their close proximity creates a single subject. The portrait is very effective even though there are two separate people that create a single main element due to their close proximity / physical overlap.

[DanK]

I would put it differently: the closer things are together, the more people are likely to perceive them as a single subject. It's a matter of probability and individual variability among viewers. I don't find the categorical distinction helpful.

Take the image below, which dates back roughly to 1890, taken by an unknown studio photographer in Lithuania. If the two in Sander's photo are one subject, are the two people in the image below far enough apart to be "two subjects"? I don't think it's a meaningful or helpful question. But by now you know my bias. In general, I don't often find rules or categorical distinctions all that helpful. It is helpful to think of these things as guidelines---e.g., if you want people to perceive two subjects, the more distance between them, the more likely that is.

[Geoff F]

For me, two people who are close together but looking in different directions would be two subjects; while two people who are slightly apart but looking in the same direction could be considered as one subject.

The vases are of a similar material and colour so they remain one subject even when apart but if one was pottery while the other was glass and were different colours they would be different subjects.

Or to put it another way, is what connects the items 'stronger' than what separates them?

[Manfred M]

This has little to do with "rules" but rather with human physiology, i.e. how humans perceive what they see.

Unlike the "rules", there has been a lot of research by psychologists, starting with Alfred Yarbus (Soviet Union) in the 1950s and 1960s, on evaluating how people see things. Yarbus and those that followed him tracked eye movement when people looked at images.

[DanK]

This has little to do with "rules" but rather with human physiology, i.e. how humans perceive what they see.

Unlike the "rules", there has been a lot of research by psychologists, starting with Alfred Yarbus (Soviet Union) in the 1950s and 1960s, on evaluating how people see things. Yarbus and those that followed him tracked eye movement when people looked at images.

Interesting. I'll have to do some reading. I'm not sure how strong the link is between saccades and how people subjectively perceive an image. In the image by Yarbus that is common on the web, you can see the eyes going back and forth between the eyes of one face, but viewers presumably see the image as one face. At any rate, in practical terms, I think it's not binary. The more one wants something to be perceived or thought of as multiple things, the more one should have some separation, either distance, color, or tone. For example, in your initial image, if one vase had been bright red and the other bright green, I suspect almost all viewers would say that they were looking at two things, not one.

In my image of callas, I deliberately used a fully black bacground, which creates clear separation between the flowers. I suspect many people would have perceived this somewhat differently if the background had been out of focus greenery, but that's just a guess.

[Chataignier]

I seem to have stirred the hornet's nest.

For what it's worth, I agree with Dan, it's VERY subjective.

Manfred quotes the Sander image of a couple very close together as showing how two become one if sufficiently close. The image in question is profoundly 'two subjects' for me because I find my eyes flicking between the two faces, or more exactly, between the eyes of the two people. Had the image been cropped tighter to show only the faces, I'm sure most people would agree with me. Had it been cropped wider to show most of the two bodies, then I imagine the reverse would be the case.

For me, your two vases remain stubbornly two Manfred.

[Manfred M]

I seem to have stirred the hornet's nest.

For what it's worth, I agree with Dan, it's VERY subjective.

Manfred quotes the Sander image of a couple very close together as showing how two become one if sufficiently close. The image in question is profoundly 'two subjects' for me because I find my eyes flicking between the two faces, or more exactly, between the eyes of the two people. Had the image been cropped tighter to show only the faces, I'm sure most people would agree with me. Had it been cropped wider to show most of the two bodies, then I imagine the reverse would be the case.

For me, your two vases remain stubbornly two Manfred.

I feel that there is certainly a great deal of subjectivity here, but also that the semantics are not clear.

Let me try another example. If we have a group shot, the subject is the group (whether that is a group of two or a group of 30). This remains true, so long as the group is reasonably cohesive and in relatively close proximity to the other members of the group.

If the larger group breaks into a number of smaller groups that are physically separated or if there are a few people that are physically scattered around that do not appear to belong to one or more of these groups, they become separate subjects. I agree that this is going to be subjective, to a point. The farther that these individuals are apart from one another or the main group(s), then they become separate subjects.

If we start thinking about composition this way, it becomes easier to either analyze an image or correct our composition to ensure that it is cohesive.

[DanK]

This is partly a matter of concepts, not just underlying psychological/neurological principles. It's how we think of a "group". Say instead you were looked at a herd of 100 elk. If they spread out to feed while you were setting up your shot, would you stop thinking of them as a "herd"? I wouldn't.

Both psychological processes and concepts help shape how we think of bunches of things. The real question is how people see the image, which is a result of both of these things and many more aspects of both the image and the person. It's not entirely predictable. For example, I would be more likely to see two vases close to each other as two subjects rather than one if they were markedly different colors, sizes, and shapes. Someone else might not. In some cases, it may not even be part of how a person views the image. For example, it never once occurred to me in working on the image of callas that it might be one, two, or three "subjects". All that concerned me was trying for an appealing arrangement (another relevant concept) and appealing lighting.

There are other cases where the count does matter. If you have 20 people standing on two close clusters of 10 with a large distance between them, most people would see this as "two groups". As you narrow the gap, an increasing number of people would see it as "one group with a gap". I think it is is partly a matter of the concept of a group, but if the gap is big enough, one also looses a sense of a primary focus.