James Gibson begins his final book by making a distinction between two senses in which the world surrounds the animal. The whole project of The Ecological Approach to Visual Perception depends upon the distinction Gibson makes between the physical world surrounding animals and the environment surrounding them. The physical world contains everything from sub-atomic particles to galaxies, but it does not contain meaning. Perception does contain meaning. If the world that animals perceive is the physical world, it is mysterious why perception is meaningful. The traditional solution to this mystery, which Gibson rejected, is that animals make the meaning somehow. For animals like us, the traditional solution has it that our brains construct the meaning in private, and that is where the meaning lives. In proposing that the world that animals perceive is the environment, not the physical world, Gibson proposes a different solution to this mystery. Perception is meaningful, on Gibsonās account, because there is meaning in the environment that animals perceive. Animals do not create meaning; they discover it in the environment. Gibsonās solution amounts to a radical rejection of the understanding of the nature of the world and its relationship to experience, perception, and knowledge that had been in place since the founding of modern science. Gibsonās solution is the key move that motivates the rest of his book, and indeed the whole of his ecological approach to psychology. The distinction has, however, created new problems. The aim of this chapter is to argue that Gibson should have made a distinction between not just two, but three different senses in which the world surrounds the animal.
This distinction is both appropriate and intuitively appealing. But it fails to include a further distinction: that between the evolutionary history of a species and the personal history of a particular individual animal. These are quite different things. We suggest that Gibsonās lumping together of these ideas has led to no small number of confusions, and to unnecessary controversy among proponents of ecological psychology.
It is necessary to subdivide what Gibson called the environment into two quite distinct concepts. We propose to replace his two-way distinction with a three-way distinction between (1) the physical world; (2) the species habitat; and (3) the animal-specific Umwelt.
A Troublesome Observation
The additional distinction we are proposing here is not novel. Gibson himself was quite aware that his use of āenvironmentā referred ambiguously to both the surroundings of an individual living animal (a token) and the surroundings of an idealized member of a species (a type), and he noted that this ambiguity could potentially cause confusion. Gibson writes (1979/2015, p. 3):
The environment consists of the surroundings of animals. Let us observe that in one sense the surroundings of a single animal are the same as the surroundings of all animals but that in another sense the surroundings of a single animal are different from those of any other animal. These two senses of the term can be troublesome and may cause confusion. The apparent contradiction can be resolved, but let us defer the problem until later. (The solution lies in the fact that animals are mobile.)
As is clear here, Gibson thought he could resolve the confusion by pointing out that animals move around. Gibson elaborates on this at the end of Chapter 3 (Gibson, 1979/2015, p. 38):
In the course of time, each animal moves through the same paths of its habitat as do other animals of its kind. Although it is true that no two individuals can be at the same place at the same time, any individual can stand in all places, and all individuals can stand in the same place at different times. Insofar as the habitat has a persisting substantial layout, therefore, all its inhabitants have an equal opportunity to explore it. In this sense the environment surrounds all observers in the same way that it surrounds a single observer.
What Gibson says here is more or less true. For most of the objects and surfaces around us, we can station our eyeballs relative to those surfaces at a point of observation that was previously occupied by one of our fellows. (There are exceptions to this: only you can see the end of your own nose from the position of your own eyeball; Gibson, 1967a.)
But this does not really resolve the tension. There are many ways in which the world can look different to two different members of the same species, even if we all keep moving around in the appropriate way. This is not because we are trapped in our own private mental world, as the traditional view has it, but simply because we are different animals with different abilities.
Take a Chinese newspaper. Neither of the authors of this chapter knows how to read Chinese. We might be able to recognize the general shape of the characters and make the judgment, āThat looks like Chinese writing,ā or again we might recognize someone in a photograph printed on the front page. But we cannot read any of the headlines. The script is not meaningful for us in the way that it is for someone who can read Chinese. Similarly, the cockpit of a plane looks different to a trained pilot than it does to a novice or to a child. The pilot can do things with the buttons and levers that non-trained individuals cannot. Or, at an even more basic level, think about climbing a set of stairs. The stairs look different depending on whether you are a toddler or a long jumper or you are someone who has just had a hip operation or you are wearing high-heeled shoes.
Whatās missing, in Gibsonās elision of the two senses of environment here, is an acknowledgment that the way the world looks to us is to some extent a result of the way we currently are as individuals. In claiming that we all share the same environment because we can in principle all see the same surfaces from the same set of positions, Gibson is treating perception as if it were an idealized process. He is suggesting that perception can be separated from the perceptionāaction loop that comprises the activity of a living animal with a history of engagement and learning. This is odd, because Gibsonās overall project throughout the book is to deny that perception is a separate phenomenon in this way, and to promote the view that perception and action are inherently one and the same process.
We suggest, then, that Gibson was right to point out the ātroublesomeā ambiguity of the term āenvironment,ā and that he was prescient in noting that this ambiguity āmay cause confusion.ā Indeed, as we will show below, it has caused confusion. The tension can be resolved, but not in the manner that Gibson proposed. Instead, we need to make a further distinction, between: the environment as it exists for a typical member of a species, a habitat; and the environment as it exists for a particular living animal, an Umwelt.
It is important to note that the three senses of environment do not define three distinct universes. Rather, the three senses are overlapping and nested. Briefly, the habitat is the physical world considered relative to a typical or ideal member of a species, i.e., it is a type of environment. It is a complementary term to a species. The habitat continues to exist even when a given animal dies. The Umwelt, meanwhile, is the physical world considered relative to a particular living individual animal, i.e., it is a token instance of a habitat. Umwelt is a complementary term to a specific organism. When that organism dies, its Umwelt necessarily ceases to exist. (The term Umwelt originates in the work of Jakob von UĆ«xkull, and refers to the world as it appears to a given animal. See Kull, 2009, for a discussion of the history of the term.)
We will not here attempt to provide watertight definitions of the three concepts of world, habitat, and Umwelt. Instead we offer a set of illustrative differences in Table 1.1. The upper part of Table 1.1 sets out some broad distinctions between the three concepts. The lower part summarizes the material below, in which we discuss how making the three-way distinction helps clarify the concepts of affordances, ecological information, learning, and the social.
Table 1.1 Some distinctions between the world, habitat, and Umwelt
| Physical world | Habitat | Umwelt |
| Structure | Resource/function | Meaning |
| Scatle-free; atoms to galaxies | Terrestrial scale | Terrestrial scale |
| View from nowhere | Ideal perspective for typical member of a species | Has a first-person perspective |
| Exists prior to any animal encountering it | Exists for typical member of a species | Brought forth through development and active exploration; enacted |
| No complementary term | Complementary to āspeciesā | Complementary to āorganismā |
| Orthogonal to life | Complementary to a form of life | Complementary to a living creature |
| Correspondence relations | ā¢ Affordance-resources ā¢ Landscape of affordances ā¢ Potential opportunities ā¢ Affordances as dispositional properties of the habitat | ā¢ Opportunities and solicitations ā¢ Field of affordances ā¢ Opportunities ā¢ Affordances as relations in the Umwelt |
| Thermodynamic information; Shannon information | ā¢ Information-about the habitat ā¢ Conventional information (in the human habitat) | ā¢ Information-for an exploring organism ā¢ Saturated with linguistic behavior (in humans) |
| History of pure structure, reconfigured by physical processes | ā¢ Has a history in evolution ā¢ Expanded by animal activities (niche construction) ā¢ Target of learning ā¢ Differentiated in learning | ā¢ Has a history in development ā¢ Enriched by skill acquisition ā¢ Location of learning ā¢ Enriched by learning |
| Orthogonal to social processes; a set of points in space | ā¢ Shared among members of a species; continues to exist when a given animal dies ā¢ Behavior settings | ā¢ Ceases to exist when a particular animal dies ā¢ Places |
Some Tensions and Their Resolutions
Replacing Gibsonās two-way distinction with the three-way distinction we propose here should allow us to re-evaluate how we think about the rest of Gibsonās book, and indeed about the 40 years of research that has followed. The entire program is predicated on treating āthe environmentā as an unproblematic fundamental. Ecological psychologists happily state that their unit of study is the āanimalāenvironment system.ā But this phrase maintains the ambiguity identified above: are we talking here about a typical member of a species residing in the environment (of that species), or are we talking about a particular animal in its environment (its Umwelt)?
As a result of this ambiguity a number of contradictions and controversies have built up over the years. The fact that we are still wrestling with what are supposedly the core concepts of the field so many years after the program was instituted suggests that the basic distinctions that Gibson made might not have been quite fine enough.
We will focus here on four key areas where the contradictions bite most acutely. We suggest that the habitatāUmwelt distinction can help resolve the tensions that have arisen in each of these cases. We focus on affordances, information, learning, and the social.
Nobody Knows What Affordances Are Anymore
First, a tension has arisen in the literature over how we should think about affordances: are they dispositional properties, relational properties, or something else?
Gibson does not get down to defining affordances until Chapter 8 of his book (see also Wagman, Chapter 8, in this volume). The way he initially defines affordances there is dependent on the distinction made in Chapter 1. āThe affordances of the environment,ā he writes, āare what it offers the animal ā¦ā (Gibson 1979/2015, p. 119). Notice that āthe environmentā is here treated as an unproblematic term. Indeed, the way Gibson begins Chapter 8 seems to imply that affordances are properties of the habitat, in our terms: they are objective features to be found āout there.ā They are a special way of describing the layout of surfaces and objects, perhaps. They are to be understood independently of any given animal, and are therefore not properties of the Umwelt. Thus, Gibson asks, āHow do we get from surfaces to affordances? And if there is information in light for the perception of surfaces, is there information for the perception of what they afford?ā
Consistent with this way of describing affordances as things that are āout there,ā Turvey (1992) attempted to provide a formal definition of affordances as dispositional properties belonging to objects, surfaces, and the like. On this account, affordances are actualized when they come into contact with an animal that possesses a complementary dispositional property, namely, an effectivity. A flight of stairs has a property, āclimbable,ā that is actualized when a stair-climbing animal comes along and climbs the stairs.
In a similar vein, Reed (1996a, p. 26) insists that affordances āare aspects of the environment of all organisms, not just features of the environment of one creature.ā Reed proposes that affordances are in fact resources, or persisting features of the habitat of a species. Because of this, affordances are able to exert evolutionary selection pressure on populations across generations.
So that is one story about affordances: they are āout thereā properties that animals can encounter. The problem is that Gibson also suggested an entirely different way of thinking...
