Feature Article


Aims in Bonsai: A Psychological Perspective

Discuss this article >>

By Dave Williams

Figure 4.

Sterotypical image of what a bonsai is expected to look like.
Bonsai and photograph by Walter Pall

This article originated as a series of posts on the BonsaiTalk forum. The original poster stated that there was a tree near his house that 'looked like a bonsai' to him and to other members of his family who had 'no real knowledge of bonsai techniques'. His question was: Why does this tree look like a bonsai more than any other tree?

I think this question raised some interesting issues concerning the relationship between trees in nature and bonsai, not least because John Naka's statement "Don't try to make your tree look like a bonsai, try to make your bonsai look like a tree" only makes sense if the relationship (and differences) between 'tree' and 'bonsai' is understood. Being a Psychologist I decided to explore this relationship from a psychological perspective.

Why do some trees in the wild look like bonsai?

I think there is often perhaps some confusion between the object (the tree) and the subject (our ideas concerning trees). When we view an object such as a tree, we are usually dealing with the projection of stereotypes or, more accurately, 'schemata'. The tree, objectively, looks like a tree. It only looks like a bonsai to someone with a schema for bonsai (i.e. a preconceived cognitive 'template' for the defining characteristics and form of a 'bonsai'). With no a priori knowledge of bonsai at all, it would just look like an odd shaped tree; perhaps a caricature of a tree.

The ability to form schemata is a mechanism that allows us to classify and categorise things quickly. It happens below conscious awareness and is essentially a time-saving mechanism. It means that when an object comes into our perceptual field, we do not have to run through our entire list of known objects in order to classify it.

When we are first exposed to a novel object, say, a tree, we will create a schema for it. After having done so, any other object that fulfils the criteria for our 'tree schema' (e.g. big, brown at the bottom, green at the top, many branches, many leaves,) will evoke that schema so we can immediately classify the object without having to consciously think about it. The more exposure we have to different objects within a category (e.g. different species and shapes of trees), the more rich and complex our 'tree schema' becomes.

Schemata are very powerful. Take for example, the 'smiley' badge (Figure 1). This unfailingly evokes our schema for a smiling face, even though it consists merely of two dots and one curved line encompassed within the circle of the badge. In no way is it a real face, nor does it present any of the real characteristics of a face (e.g. it is usually bright yellow and has no ears or nose) yet it presents representative characteristics that are so evocative of our 'face' schema we cannot help but perceive it as a smiling face .

Figure 1.

Figure 1. An example of an arrangement of basic elements (left) which evokes the schema for face (right).

Whilst the ability to form schemata is an extremely useful faculty that allows us to classify novel objects quickly, it has a down side. It means we do not have to think about what we are seeing and so can result in stereotyping. Herein lay the crux. In the same way as we can form a schema for 'tree' (Figure 2), we can also form a schema for 'bonsai'. This schema can be separate and disparate from our 'tree' schema' having its own particular set of trigger criteria. In effect, it is a stereotype (Figure 3).

Figure 3.

Figure 2. An arrangement of elements (circles and wedge) that when presented in silhouette evokes a schema for tree, but not for bonsai.

Figure 4.

Figure 3. An arrangement of 7 curved lines (left) that evokes the 'stereotype' schema for bonsai, but not for tree.

Having been exposed (probably many times through various media) to images of 'bonsai', often stereotypical in themselves (e.g. Karate Kid et al.), we develop schemata for the stereotypical physical attributes that make up a 'bonsai'. Then, if we see a few or more of these attributes reflected in a tree in the wild, it simply evokes our schema for 'bonsai' and the label is applied automatically. This is why some trees in nature 'look like bonsai'. They evoke our 'bonsai schema' rather than our 'tree schema'.

Because of this, I feel the natural tendency develop schemata is often more of a hindrance than a help, particularly when it comes to something as subjective as bonsai, because it has implications for the way we approach bonsai design.

Bonsai means literally 'tree [in a] tray' or something close (I don't speak Japanese), but, in my opinion, that's all it should mean. However, to many, the word 'bonsai' actually evokes vivid cognitive representations (schemata), where the English term 'potted plant' evokes no such vivid and concrete preconceptions.

Many who come to bonsai do so having developed such stereotypical schemata and so have certain a priori preconceptions concerning what a bonsai should be (i.e. what specific characteristics a plant needs to possess in order to 'qualify' as 'a bonsai'). Then, if they're anything like me, they will waste a lot of time initially in the (often brutal) process of trying to force a plant to conform to these preconceived notions of what it 'should be'.

In my experience, real learning begins with 'unlearning' what one believed one already knew (in this case, the schema evoked by the term 'bonsai'). I think to view a tree in nature and think "That tree looks like a bonsai" is evidence for the existence of these preconceptions. It would perhaps be better to be able to view a tree in nature and think "that tree has a pleasing form that I might like to reflect in my bonsai".

Bonsai are designed to evoke tree schemata.

A common belief concerning bonsai is that the aim is to grow miniature trees in the form of full sized, mature trees. However, I don't think we do grow miniature trees in the form of full sized trees. I think it's more true to say that we grow trees that represent full sized trees (i.e. miniature representations of trees that evoke a viewer's schema for full sized trees).

Obviously, if we concentrated on size alone and just attempted to miniaturise trees to scale, it would present insurmountable problems (e.g. leaf-size, bark texture and so-on). So, we concentrate on proportion and perspective rather than scale and use visual tricks to confound the viewer's subjective senses of proportion and perspective.

One example of such visual trickery is defoliation. This is done to increase fine ramification and to reduce leaf size and also increases the overall number of leaves (due to the increased ramification). But the aim is not to try to reduce the leaves to scale. There is no way anyone could ever expect to get leaf size to scale on any bonsai. But that's fine, because that's not the objective. The objective is to present a few of the attributes that evoke the viewer's schema for 'tree'.

Two of the defining characteristics of a tree in nature is that, in general, it has 'many' leaves and a clearly defined outline (silhouette). In bonsai, defoliation decreases leaf size and increases ramification and overall number of leaves which together fulfil these requirements of the 'wild tree' schema. Thus, the leaves on a bonsai don't have to be to scale, they just have to be small enough to define an outline and numerous enough to suggest 'many' in the mind of a viewer. Larger leaves break up the outline and so break up the illusion. Even one large leaf on an otherwise well groomed bonsai look incongruous and makes it very hard for a viewer to accept the illusion, as those who have ever grown a sacrifice branch on a more mature bonsai will know.

Another example of visual trickery is tricks of perspective. In bonsai, whilst the (much) wider base of the trunk may move away from the viewer, the thinner, higher portion then sweeps forward towards the viewer. If one stands in front of a large tree with a straight trunk in the wild, as one's gaze moves up the trunk, the same thing appears to happen. The base appears a lot wider than the trunk higher up (even though the trunks of wild trees are often parallel) and the higher portion appears to loom forward from one's own point of view. This is a trick of perspective that, when physically recreated in a small tree, suggests a much larger tree.

Along with tricks of perspective, like caricaturists, bonsai artists try to enhance those stereotypical elements in a bonsai that suggest size, massive proportion and age; a trunk taper that would rarely be found in nature (but that due to perspective makes the tree appear large), a gnarled bark, downward sweeping branches, rounded apices and so-on; characteristics that are found only in older, larger trees.

Bonsai really is an art of illusion. These tricks of perspective, proportion and elements of caricature are used to make a relatively young and comparatively very small tree appear much older and larger. These techniques are basic to most and come even before other techniques peculiar to different styles and that suggest the life story of the tree, and evoke the environment surrounding the tree (cliff face, mountain top, open field, windswept ridge and so-on).

The end result is that by presenting in artful harmony the basic representative characteristics of a tree in nature, good bonsai aims to evoke a viewer's schema for 'tree' rather than their schema for 'bonsai'. A bonsai is a subjective representation of a tree that allows the viewer to suspend easily objective disbelief and to slide easily into the illusion of a tree in nature.

The same principle applies to 'flaws' in bonsai such as crossing branches, inward growing branches, inverse taper and so-on. Why is it bad practice to have crossing branches?' Because crossing branches in bonsai 'jar' the eye and break the illusion. A large tree in nature can have crossing branches and be beautiful. But a large wild tree is large, and is not trying to represent anything other than what it is. It simply 'is' (and therein lies its beauty). However, a bonsai is trying to represent something. Moreover, it's a lot smaller. Even in trees in nature, crossing branches, frog-leg trunks and so-on can appear 'ugly' to human sensibilities if these 'flaws' are large enough to impinge upon the overall impression. These elements make much more of a visual impact in bonsai than they do in the wild as bonsai are significantly smaller.

As the idea of bonsai is essentially to get a viewer to 'buy into the illusion', we aim for flow and harmony. Primary, secondary, tertiary and quaternary branches all flow along the same lines away from the trunk line. As long as the eyes of the viewer are allowed to flow with them, the illusion will hold. Where one branch or twig suddenly changes direction and moves towards the trunk, the viewer's gaze is halted and the illusion is broken.

There is arguably no such thing as a perfect bonsai (it's too subjective for 'perfection' to be a realistic objective anyway), but a lot of the cunning and skill in creating bonsai is involved in hiding, including or, in some other way dealing with the flaws inherent in most bonsai in such a way that they do not jar the eyes of the viewer and break the illusion. Thus, arguably, any bonsai in which the illusion of what it is trying to represent holds together for each viewer is a perfect bonsai.

The importance of developing schemata.

The complexity of schemata is determined by environment and learning. That is, the degree of exposure to different objects within the same category.

Even expensive wine tastes awful to me; like vinegar that hasn't quite made it. It's not the fault of the vintner, it's down to my lack of exposure to wine. Preferring, as I do, to suck down lots of tequila I simply never developed the complex schema for wines that is a necessary element of the vintner's skill. My lack of appreciation says absolutely nothing about the quality of the wine or the vintner. However, if I was a vintner and wine tasted awful to me, I would have a real problem.

In the same way, a person's schema for tree depends upon the person's exposure to trees (I mean actually noticing and paying attention to them). This of course means that a person's schema for trees in nature will be strongly influenced by their environment. A person growing up in flat, open farmland will develop a different schema for 'tree' than a person growing up in a windy, mountainous region, because the types and forms of trees in each environment will be different.

In some cases however, certain people (particularly city dwellers) will have little exposure to trees in nature at all, and when they go out to the country, will not really notice trees, beyond the 'oh, there are some trees' experience. Many people I know can't tell what species a tree is, or even one species from another. They're all just 'trees' (big, brown and green, lollipop shaped things growing at the sides of the roads). In such cases, the schema for 'tree' will be very simplistic and quite fixed, lacking flexibility and the capacity to accommodate the more unusual examples of trees that can be found in nature (this can actually be seen if you get such a person to draw a tree).

Similarly, in some people, the only exposure to bonsai they will have had will have been to stereotypical examples (cf 'Karate Kid'). In which case, they will have a similarly simplistic and inflexible schema for 'bonsai' that is likely to be very different from, and incongruous with their schema for 'tree'. In such people it's highly likely that many excellent bonsai will not match their schema of 'real trees', but then (ironically) it's also likely that many excellent bonsai won't match their schema for 'real' bonsai either.

I don't think it's catastrophic if our bonsai do not evoke the schema for tree in all viewers if those viewers have little experience of trees and no experience of bonsai. It's enough if they consider them pleasing to look at. We can only expect a bonsai to evoke a tree schema in a person that has a comparatively broad tree schema to begin with and that only comes from exposure (conscious experience of different types and shapes of tree). However, if we wish to create bonsai, a rich and complex schema for 'tree' is an absolute necessity.

This of course begs the question; what then is the best inspiration for a bonsai, outstanding existing bonsai or trees in the wild? I would say both, but neither alone is sufficient.

To use only existing bonsai as inspiration puts one in the 'bonsai schema' trap in which one begins to form rigid preconceptions of what constitutes a 'good bonsai'. Whilst an outstanding bonsai is a distillation of the most rewarding characteristics of trees in nature, to use such bonsai solely as one's inspiration would lead to copies of copies of copies. Each may be a more 'pure' distillation of the rewarding characteristics of the bonsai preceding it, but then each would also be one step further removed from the reality of 'a tree'. It wouldn't be long before bonsai lost completely their 'tree-ness'.

On the other hand, to use only trees in nature for inspiration would mean reinventing the wheel over and over again. To produce in a pot a miniature replica of a natural tree would provide very little reward. It would appear to be completely out of proportion because much of what impresses us about large trees in the wild is produced in bonsai using the visual trickery described above.

Without these tricks, a tree would (generally speaking) have a relatively thin, parallel trunk with little or no taper or movement. It would present a solid wall of foliage that obscured any view of its structure. It would also contain all those elements that 'jar' the eyes; crossing branches, branches pointing at the eyes of the viewer and so-on. Due to the presence of these elements and its small size, it would be a dilution, rather than a distillation, of the most rewarding characteristics of a tree in nature.

To gain any reward from these bonsai, we would have to learn from scratch how to represent in miniature through manipulation of perspective and other tricks, those characteristics that we find most rewarding about trees in nature. Why bother when generations before us have already done so?

However, a combination of existing bonsai and natural trees as inspiration works well in my opinion. Trees in nature can provide novel and fresh inspiration that is in sympathy with the 'natural' tree schema we have developed as individuals (shaped by our individual environments and experiences), and which we wish to represent in our bonsai, but existing good bonsai provide real life examples of how rules and techniques are applied in order to best represent our inspiration. That is, how best to fool the eyes of a viewer into perceiving a large and mature tree and the suggestion of its immediate surroundings, and to ensure that what is perceived is rewarding.

More importantly, using both trees in nature and existing bonsai for inspiration inhibits the formation of separate and distinct schemata for 'tree' and 'bonsai', merging them into one rich schema which includes the most rewarding elements of both; the visual appeal of trees in nature and the artful trickery of bonsai.

Trying to make your bonsai to look like a tree.

John Naka said, "Don't try to make your tree look like a bonsai, try to make your bonsai look like a tree". But, if a bonsai is a representation of a tree (rather than a replication in miniature), then won't all bonsai look like bonsai rather than trees? I think this is simply a problem of semantics. By definition, all bonsai look like bonsai. However, not all bonsai look like trees.

On the one hand we have the bonsai as an artful representation of a tree in the wild. The rules and techniques have been applied to create the illusion of a venerable tree and its surroundings. On the other hand we have bonsai as an assemblage of the stereotypical features of other bonsai which is a result of falling into the 'bonsai schema' trap.

Let me illustrate by way of (hypothetical) thought processes involved in the creation of each:

On the one hand:

I have a tree in training. I look at it and imagine its surroundings. I think this one is growing in pasture in East Anglia (very flat, but good growing land). There is plenty of rain and the soil is rich. There is a constant (but not severe) prevailing wind because the land is so flat.

So, the tree will appear healthy and lush. The trunk will have a gentle movement that comes with age, but will generally tend to move gently away from the wind. The branches will be slightly longer on the lee side than the windward side. The lower branches will slope down gently under their weight because the tree is so 'old', but the longer branches on the lee side will slope slightly more so because they have 'always' been heavier.

Another example; I have a tree in training and again I look at it and imagine its surroundings. I think this one is growing in dry mountain gravel high on a slope. It has poor soil, not much rain and strong, grit-laden winds coming up the slope.

So, the tree will be stunted. It won't have much foliage because it can't support much. The trunk and branches will be twisted away from the wind but the branches won't slope gently down because the foliage was never heavy enough to pull them down. Over time, the wind and grit will have blasted the bark from the trunk and branches on the windward side. What little foliage it can support is growing in the lee side, gaining some shelter from the shade of its own trunk.

Each of these trees tells a story of the life of the tree. In each case, the story is coherent and makes sense. There is nothing there to jar a viewer awake from the illusion.

On the other hand:

I have a tree in training. I look at it and invoke my 'bonsai' schema. What features do I need to include in order for this tree to 'look like' a bonsai? Ok, well, it needs a bendy trunk (bonsai have bendy trunks). It also needs the first branch to be the biggest and coming from one side. I'll add a jin (I know bonsai have jin, I've seen them). I have a dremmel, so I'll create a shari so I can carve the trunk (I've seen that in bonsai too). It should have small foliage pads and branches that bend down and so-on.

The end result may possibly be visually pleasing, but the story it tells is incoherent and full of visual non sequiturs. Why does the trunk move like that? What bent it? Why is the shari on the same side as the longer branches? Surely the side subject to the harshest conditions should have the shortest branches? Why are the branches bending so much under the weight of so little foliage? Why does the apex move toward the shari if that's where the wind is coming from? Why is the jin right above the biggest, most healthy branch on the tree and on the opposite side to the shari?

What we have is a 'stereotype' of a bonsai; a tree that contains all the stereotypical features of a bonsai but in a combination that makes no sense and raises too many questions to allow a viewer to slide easily into the illusion. All these examples will look like 'bonsai', but some will make sense and others won't. This is what I think John Naka meant. I believe he was arguing for the former approach above and I think it's one of the most profound statements I've ever heard in bonsai, because of its implications for bonsai design.

No bonsai exists in isolation as 'just' a tree. Each tree carries with it the story of its life and thus, the suggestion of its surroundings and environment. If the story it tells makes no sense, the illusion is shattered to the detriment of the tree itself.

The upshot is that in order to develop the rich tree schema tree we need in order to create effective bonsai we can't just look at trees. We have to look around them at where and how they are growing. We need to understand the forces that create the most rewarding characteristics of those trees in the wild, in order to be able to combine the representations of these characteristics in a way that makes sense and tells a coherent story.

For example, why don't we create shari on deciduous trees like maple? Well, because deciduous trees rarely grow in situations where the necessary forces for natural shari exist (e.g. high, drought ridden rocky slopes with strong, grit laden winds). We never see such examples in nature, so what would such a bonsai represent? The 'story' told by the shari would be incongruous with the nature of the tree. The only time we see shari on deciduous trees is in the presence of disease and nobody wants to copy those. They are not rewarding to look at.

The purpose of bonsai.

The purpose of a bonsai is simply to provide reward. Note that here I'm using the term 'reward' in its basic, psychobiological sense. Reward is an emotional response to a stimulus. It is mediated by the limbic brain and results in a positive affect (emotional state) and the behavioural motivation to orient towards and approach the stimulus. Reward is an extremely powerful reinforcement for memory and behaviour (addiction is based upon the artificial excitation of neurological reward centres).

The human senses and memory are quite lazy and are very easily fooled. We say that there are many beautiful trees in the wild, and this is true, but there are many more that are not so beautiful (i.e. not rewarding to look at). However, we don't tend to notice or remember the not-so-beautiful ones. They don't 'stand out' and so they get consigned to long-term memory without recall traces along with so many other mundane features of our immediate environments, even though we might see them every day (what colour is the front door of the house directly opposite yours? How many windows does the front of that house have?).

We tend to notice, and so remember, the more striking trees and those that we find rewarding to look at but which, arguably, are the exceptions to the rule. Those trees that we find rewarding to look at are the trees we are trying to represent in bonsai, simply because they are rewarding to look at. Indeed, our initial evaluation of a bonsai as 'good' or 'bad' is based upon the immediate feelings of reward we get from looking at it.

If we accept that the aim of bonsai is not to create a miniature replica of a tree in the wild, then it follows that the 'rules' of bonsai were not formed with that aim in mind either. The 'rules' of bonsai have been formed over hundreds of years with the single aim of identifying the most rewarding characteristics of trees in nature. For example, the trunk proportion 'rule' in bonsai (i.e. 6:1 height/calliper) is a manipulation of perspective designed to fool the viewer into the illusion that they are standing close to the base of a large trunk that is growing up and shrinking towards vanishing point in the distance. The reason for the rule is simply that looking up at a magnificent mature tree is very rewarding, whereas looking at a little sapling is less so.

The rules show us how we may represent such characteristics and combine them harmoniously in such a way that looking at it evokes our schema for 'tree in nature' and so produces the same rewards, or an more pure 'distillation' of them, in trees small enough that we can have many of them, close to us in our own gardens.

There is a current movement towards more naturalistic forms in bonsai led by people like Walter Pall. I have absolutely nothing against this at all. I quite like the natural forms of trees I see in the English countryside. Nonetheless, even Mr. Pall will still utilise 'rules' in the development of these naturalistic trees. To renounce all such rules completely would result simply in free growing saplings, shrubs or stumps in pots. In order for the trees Mr. Pall creates to be rewarding to look at (and they are), some basic rules of form must have been followed.

Irrespective of whether we choose to follow established rules, or abandon them and find our own way (which, in my opinion, is entirely a matter of personal choice), the success or failure of a bonsai boils down to one thing; reward. If a bonsai is not rewarding to look at then, as a bonsai, it is utterly pointless. It does not serve the purpose for which it was created. If a bonsai represents well the rewarding characteristics of trees in nature in a way that tells a coherent story and evokes a viewer's schema for 'tree in nature', then it will be rewarding to look at and thus, a success.

The more naturalistic trees designed by people like Mr. Pall are rewarding to look at because they successfully evoke the 'tree in nature' schema in viewers. However, they may not necessarily evoke the 'bonsai' schema in some viewers. But then, that's not an issue because, as argued above, that's not the objective. In this, I believe the statement by John Naka applies as much to viewers of bonsai as it does to growers of bonsai (to paraphrase): Try not to look at a tree and see a bonsai, try to look at a bonsai and see a tree.

Discuss this article >>