Affordance Gibson PDF

Journal for the Theory of Social Behaviour: r g : ~
0021-8308 82.50
Affordances and the Body:

An Intentional Analysis of Gibsons
Ecological Approach to Visual Perception
HARRY HEFT
The concept of affordance is one of the most controversial and debated features
of James J. Gibsons ecological approach to perception. With this concept
Gibson offers a new approach to a knotty problem in perception theory;
namely, the problem of accounting for meaning in perceptual experience. The
controversial nature of the affordance concept stems largely from the fact
of its growing out of an approach to perception that is a significant departure
from standard formulations. The aims of this paper are two-fold. The first
is to provide a detailed analysis of the concept of affordance, and in the
process to draw out some of its implicit theoretical claims concerning the
nature of perception. This analysis will be presented in the first part of the
paper. Exploration of the implicit assumptions of the affordance concept will
reveal the underlying intentional character of the ecological approach to
perception.
The second aim of the paper is to consider the applicability of the affordance
concept to features of the human world whose meanings are sociocultural in
origin. As will be clear shortly, affordances seem most plausibly applied to
features of the environment that have species-specificor transcultural significance. However, the concept of affordance has sometimes been applied to
features that have significance only within a particular sociocultural context.
These applications have been a source of some debate. I will argue that this
extension of the affordance concept is warranted once affordances are carefully
grounded in an intentional analysis of perception. Moreover, the resolution of
this controversy will provide the basis for maintaining the broad and
fundamental character of affordances in perceptual experience. Toward these
ends, we will examine in parts two and three of the paper some issues concerning
intentionality in perception and its relationship to Gibsons ecological theory of
perception.
AN ECOLOGICAL APPROACH TO PERCEIVED MEANING
There is a need to study the perception of surfaces with a realistic attitude as well as a
phenomenological attitude. (Gibson, 1979b, p. I 12)
If we adopt a phenomenological attitude to perceiving, it is apparent that

features of our world have meaning for us. At the simplest level, environmental
features are often experienced as attractive, positive, and alluring, or inversely
as unattractive, negative, and repelling. Moreover, environmental features are
often experienced with respect to their functional significance: we perceive
features in terms of the ways we can interact with them. In short, the features of
our world are not value-free. Dewey (1934) puts it this way:
The live animal does not have to project emotions into the objects experienced. Nature is kind and
hateful, bland and morose, irritating and comforting, long before she is mathematically qualified or
even a congeries of secondary qualities like colors and their shapes. (p. 16)
How do we account for this dimension of perceptual experience?
According to the standard analysis of perception, meaning is imposed on

sensory input by mental processes. Meaning is not inherent in sensory input, nor
can it possibly be, because the basis for perception, according to this view, is
physical energies that stimulate receptors. Owing to their physical nature, these
stimuli themselves cannot convey the meanings of objects, but only, e.g., their
intensity, frequency, position, and movement, via light rays reflected from their
surfaces. I
In Gibsons ecological theory of visual perception, perceiving is not based on
stimulation of receptors by physical energies, but on the pick up of higher-order
information in reflected or ambient light by a perceptual system. (For a detailed
account of Gibsons ecological approach, see Gibson, 1966, 1979a; Reed and
Jones, 1982.)In brief, Gibson and his followers have attempted to describe the
ways in which light is structured as it is reflected from surfaces of objects, both
animate and inanimate. Their focus is on the patterns of reflected light,
particularly the invariant structure that is preserved across transformations of
the array. This higher-order structure is available in the ambient array oflight
to be sampled by a perceiver.
What is conveyed by this higher-order structure in ambient light? Formal
properties are such as object size, shape, and motion, and in addition, the
functional significance or meaning of objects for the perceiver. Let us take a
relatively simple example to illustrate the latter possibility of the functional
significance of environmental features carried in reflected light. There is a
very specific pattern of information perceivable at the edge of a precipice:
Affordances and the Body
as a perceiver stands at the edge of a cliff, the slightest movement results

in a shearing of the texture of the ground on the surface below by the edge
of the surface of support (E. J. Gibson, 1969). This transformation in the
reflected light - one which is generated by a moving observer in relation to
this arrangement of surfaces - conveys information that there is a break in
the continuity of the supporting surface; it means here is a falling-off
place. This meaning is carried in the structure of reflected light. It is a
perceivable ecological fact, not a mental construction that is imposed on sensory
input.
The concept of affordance refers to these perceivable functional meanings of
objects and events that are carried in the structure of ambient light. Although
the stimulus information for the affordance qualities of environmental features
have only been worked o u t for just a few simple cases, [e.g., graspable objects
(Hallford, 1984); looming objects (Schiff, 1965); climbable steps (Warren,
1984); seats (Mark, 1987);doorways (Warren and Whang, 1987); also see E. J.
Gibson, 1982; E. J. Gibson and Spelke, 19831, the assumption is that, as a rule,
the functional meanings of environmental features are available to be perceived
in the ambient array of light.
Affordances defined
Gibson (1979a) defines affordances most directly in the following way: The
affordances of the environment are what it ofers the animal, what it provides
or furnishes, either for good or ill (p. I 2 7 ) . The affordances of a given place
in the environment establish for an individual what actions are possible there
and what the consequences of those actions are. For example, a surface of
support at approximately knee height to the individual affords sitting on. A seat
is a feature of the environment specifiable in terms of properties of the object
(i.e., it has a particular mass, height, and width); but its parameters as an
affordance are delimited with reference to a specific individual of a particular
weight, leg length, and girth. As a result, what constitutes a seat (or affords
sitting-on) will vary among individuals with significantly different body scaling.
The relative nature of seat affordances can be illustrated by the fact that a
surface perceived as a seat by a young child may not be perceived as such by an
adult. Specifically, a foot stool may be perceived as a seat by a child and not by
an adult as a function ofleg length, and a cardboard box may be perceived as a
potential seat by a child but not by an adult because of their differences in
weight.
Affordances, then, are properties of the environment taken with reference to
an individual. As such, they have both objective and subjective qualities.
They are objective in the sense that they are facts of the environment;
what constitutes, e.g., a seat, depends on the physical characteristics of an
object .
Harry H 4 t
What [an environmental feature] affords the observer is determined by its material substance and
its shape, size, rigidity, motion, etc. What it means and what it is are not separate, as we have been
led to believe. And the observer who perceives the substance and the surface of anything has
thereby perceived what it affords. (Gibson, 1972, p. 410)
However, affordances deviate from a strict meaning of an objective property

in that they are not specifiable independent ofan individual, as are physicalistic
properties such as mass and extension. Because affordances implicate a
particular perceiver, e.g., his relevant body-scaling characteristics, they may be
considered to have subjective qualities. But affordances are not subjective in
the sense that they reside in mind, as we have seen; they are ecological facts.
Thus, affordances do not fit neatly into either of these two ontological
categories; instead they are relational in nature.
In the present context of an analysis of the environment and the individual,
relational may be set in opposition to dualistic. In a dualistic approach,
one draws conceptual, and often ontological boundaries around each domain,
separately examines the essential properties of each domain, and considers the
mechanisms by which each influences the other. This approach is characteristic
of much analysis in psychology, and it is an approach that Gibson explicitly
rejects. In contrast, a relational approach focuses on properties that exist only
when one simultaneously considers two (or more) entities. Put differently, a
relational approach assumes that (no one [constituent of an inquiry] can be
adequately specified as fact apart from the specification of other constituents of
the full subjectmatter (Dewey and Bentley, 1949, p. 1 2 2 ) . With respect to
environment-individual analysis, a relational concept refers to a property that
emerges out of the interaction between an animal and the environment.
Affordances are located at this boundary; they are synergetic properties of an
environment-animal system (Shaw and Turvey, I 981). Finally, relational
properties are not subjective entities; as Ben-Zeev ( 1984) succinctly states:
Relational is not opposite to real; it just refers to a certain kind of reality
(P. 79).
Because of the inappropriate tendency of investigators to assimilate Gibsons
theory to a dualistic framework (see Heft, 1980, 1982;Reed and Jones, 1981;
Wilcox and Katz, 1 9 8 1 )it~ may be worthwhile to look more closely at the
relational approach, which is reflected in the affordance concept.
The relational approach to enuironment-animal interaction
By now it is commonplace to point out that a living entity, unlike a physical
entity, is an open system, meaning that its continued existence and growth are
based on an on-going and reciprocal exchange with its surround. Although the
animal and the surround may be examined independently, certain critical
psychological processes are most clearly revealed within the conceptual
region that analyses ofanimal and environment mutually share. That is, some
of the most psychologically significant phenomena are relational in nature.
The hallmark of an entity with a relational quality is that its specification
implies a second entity. Many of an animals structural and behavioral
characteristics have this quality of pointing beyond themselves to implicate
particular characteristics of the environment. For example, the sensory
sensitivities of an animal bracket some types of physical energy in the
environment, among the variety ofenergies that can be identified, as well as the
range of energy within those categories to which the animal is responsive. So
humans, like all terrestrial animals, are sensitive to electromagnetic radiation,
and humans in particular, (and unlike, e.g., some insects) are responsive only to
wavelengths between approximately 400 and 700 nanometers. Thus, a
description of human visual sensitivities implies a specification of some portion
of the total environment, and conversely that particular portion of the
electromagnetic spectrum has significance because of its relation to human
vision. That is, both factors are relationally specified. The same analysis can be
performed for other aspects of the animals structure, such as other receptor
sensitivities.
Likewise, the behavioral capacities of an animal co-implicate particular
characteristics of the environment. For example, the behavior ofgrasping in an
animal means that detached objects are potentially significant features of the
environment (if, that is, the object is smaller than the handspan). Further,
sitting co-implies places that afford sitting-on, striding co-implies places that
can be stepped across or stepped up onto, and so on. Each ofthese actions is only
possible with respect to environmental features which permit them, and these
particular features are significant in relation to these actions, i.e., each factor is
re1ationally specified.
Many characteristics of an animal do not have this relational quality. Most
neurophysiological features, as well as hypothetical mental operations, can be
described without simultaneously implicating features of the environment. For
example, biochemical characteristics of cell receptors sites or hippocampal
functioning, on the one hand, or the functional characteristics of hypothesized
memory processes, on the other hand, can be considered independently of
environmental conditions. Therefore, only some aspects of an animals
functioning can be characterized as relational, and these relational processes
should be the particular province of psychology, (or at least an ecologically
oriented psychology), as opposed to neurophysiology or cognitive science,
respectively .
It can be seen then that specification of certain structural and behavioral
characteristics of an animal implicates and delimits that portion of the total
environment comprising the animals (or the species) econiche. These characteristics of an animal have a relational or transactional quality in that they
imply a second entity; they point to the surround, and together both sides
Harry H g t
express the functional compatibility existing between an animal and its

econiche. Thus, the environment, in an ecological sense, is fundamentally
relational.
Considering affordances more specifically, they too are relational in nature.
They are the environmental counterparts to the animals behavioral potentialities. Returning to the examples used previously, objects that are smaller than
the handspan (but not too small) are the environmental counterparts of the act
ofgrasping; they afford grasping. The act ofgrasping is only comprehendible in
relation to a thing which may be grasped. Objects that are surfaces ofsupport,
relative to the mass of the individual, and that are at approximately knee
height, are the environmental counterpart of the act of sitting; they afford
sitting-on. In a sense, affordances complete the unity of the behavioral act. The
affordance and the related behavior together specify goal-directed action.
Considered individually, an affordance or a specific behavioral act is but one
facet of an intentional act (see below). Each facet taken separately necessarily
implicates its counterpart. The relationship between an affordance and a
behavioral act is characterized by a mutuality, a compatibility, and a fittedness
(Turvey and Shaw, xg7g).*
Finally, the kind of relational approach presented here does not lend itself to
the standard dualistic, causal analysis typically applied to environment-animal
relations. The standard dualistic analysis draws a sharp, and often ontological
distinction between the environment and the animal; and it views causality
solely in mechanistic terms, with the directionality of the influence, that is the
antecedent determining influence, in most cases originating in the environment
and impinging on the individual. For example, the standard account of visual
perception begins with a physical analysis of the stimulus object in terms of
wavelength and intensity of light rays, which impinge on the visual receptors,
initiating neural activity, and ultimately, after considerable processing,
resulting in a percept. The particulars of this formulation vary considerably
among theorists, but this essentially mechanistic account of causality is
pervasive in the study of perception and cognition.
In contrast, the relational approach points to reciprocal or mutual influences
in the on-going and synergetic transactions of the environment-animal system.
A closer examination of the issue of causality in the analysis of animalenvironment relations will give us a clearer insight into the nature of the
affordance concept.
Causality, environment-animal relations, and affordances

To digress briefly, causality classically encompassed more than a mechanistic
explanation based solely on antecedent occurrences or factors. The identification ofcause with mechanical causality is a post-Renaissance phenomenon, and
it reflects the growing hegemony of the physical sciences for whom such an
7
explanation proved to be very fruitful and perhaps sufficient as well. The
fruitfulness and, more particularly, the sufficiency of mechanical explanations
for biological phenomena may be another matter.
Classically in the Aristotleian framework, causal categories were invoked,
in part, to contribute to our understanding of a things nature. For Aristotle,
the explanation ofsome occurrence is attributable to: ( I ) the material character
of the features involved (material cause), ( 2 ) their structural character (formal
cause), (3) their origins (efficient causality), and (4) their immanent or
teleological character (final cause; see Grene, 1963).These categories, and
those of formal and final causality in particular, emphasize the distinctiveness
of natural things. In contrast, it is characteristic of post-Renaissance science
to apply mechanistic explanation to all entities regardless of their nature.
Mechanistic causal explanation, often in combination with materialism, seeks
to reduce the diversity of natural things to a common mode of description,
if not a common level of analysis. Thus, Newton offers a grand theory of
physics applying mechanistic explanation to events in the heavens as well
as on earth; in the same vein, Locke and those who follow broadly in his
approach propose a Newtonian analysis of mind. The Lockean model has
profoundly affected psychological analysis to the present day. Behaviorism, and
its contemporary versions, e.g., information-processing, are transpositions
of this approach to observable actions and inferred mental operations,
respectively .
One might argue, however, that in order to understand the cause of some
event, it is necessary to take into account the distinctive nature of the entities
involved. Otherwise, all entities are treated identically, and their substantive
differences are ignored. This is not a problem if a reductive, materialistic
approach to all phenomena is adopted. If, however, one minimally distinguishes between physical (closed) and living (open) systems, the nature of
the entity in question needs to figure quite prominently in a causal analysis. We
may need to adopt a broader view of causality (although not necessarily the
above classical categories in particular) precisely because the specific qualities
ofliving systems, and their relationship with their surround, are fundamental to
an adequate explanation of psychological phenomena.
In what way is a causal analysis in a relational view different from that in a
dualistic approach? Let us consider two differences, and then apply this
discussion to our consideration of the affordance concept.
I ) As applied to psychology, a dualistic approach suggests that events stop
and start in a discontinuous series ofjerks (Dewey, 1896).The animal, in effect,
is in a state of relative stasis until some environmental occurrence stimulates or
goads it into activity. This admittedly sketchy account also applies to many socalled active models of perception and cognition that employ schema as a
central concept. With few exceptions (e.g. Neisser, 1976), environmental
occurrences are seen as initiating psychological activity. I n this view, the causal
Harry Hdt
relationship between environment and animal is not substantively different

than the relationship between any two physical objects.
In the relational view, environment-animal transaction is an ongoing,
largely unbroken process. There is a continuous stream of behavior or a
stream of transactions between environment and individual. Breaking into this
stream oftransactions at any particular point to find an environmental cause for
some specific behavior is not only typically arbitrary but also potentially
misleading. It is arbitrary in the sense ofdetermining when in time to look for an
antecedent cause ofa behavior, and it may be misleading in that it suggests that
cause and effect can be limited to a specijic environmental event and a specijic
behavioral act.
An example will clarify these points. Somewhat early in his investigations of
the ecology ofchild development, Barker ( 1968; Barker and Wright, 1955) and
his colleagues looked to see if antecedent events accounted for a childs
behavior. In one case described in detail (Barker, 1968, pp.1&151), Barker
observed the behavior of a child named Maud in a drugstore. The sequence of
actions the child expressed were noted in detail, as well as the social occurrences
in the childs proximity. Barker had assumed that Mauds separate behaviors
would largely be responses to discrete social inputs; that is, that behaviors
would follow from antecedent events, in the manner of stimulus-response
occurrences.
We expected behavior episodes to march along single file preceded by inputs from the environment
and terminating in outputs to the environment, as they do when psychologists are operators in
psychophysics experiments. . .. We expected to be able to predict with some accuracy from
ecological inputs to behavioral outputs. But we were wrong. (p. 147)
In fact, only about one-third of Mauds actions corresponded to immediate

social inputs. If one were to base ones judgment of the predictability of
her behavior from this analysis, it would have to be concluded that her
behavior was largely unpredictable. And yet it was not. Mauds behavior
was appropriate to the setting that she was in, and in turn, predictable in
relation to that factor. Behavior appeared to be constrained, or Lcoerced,
by the collective social forces that give rise to the setting. This observation
was one of several that led Barker to postulate the concept of behavior
setting as an important higher-order ecological phenomenon (see Barker, I 968,
pp. 151-166).
To generalize from Barkers data, since a behavioral act is not always
predictable from immediate environmental inputs, when to look for a cause of
behavior in situ may be arbitrary. Behavior does not typically correspond to a
single environmental event at a moment in time (although it sometimes may),
but rather it is compatible with, or constrained by, the environmental setting.
There is an overall fittedness between behavior and the immediate environment. I n the process of looking for an antecedent event, one is apt to overlook
Affordancesand the Boaj
9
higher-order, relational factors, such as behavior settings and affordances,
which may be in fact the more valuable explanatory constructs.
2 ) In addition to the on-going nature of the transactions within the animalenvironment system, causal influences are reciprocal, with the impetus of
fluctuations in the on-going behavior stream having its source in the
environment facet OT in the individual facet of the transaction; and this
reciprocal exchange is cumulative in its effects.
Considering the environmental side of the exchange first, occurrences or
features of the environment can lead td the initiation of a series of subsequent
transactions, as in the case of a telephone ringing in my home, and its
transactional aftermath. As with this example, most instances ofenvironmental
conditions initiating a subsequent series of transactions are likely to be intrusive
in nature (e.g., an unexpected sound or sight in the perceptual field). Further, it
is possible that the impact of particular environmental conditions is not
immediate, but rather becomes manifested sometime after the individual has
been exposed to them (e.g., the effects of some conditions in the home on child
development, Wachs and Gruen, 1982). In addition, environmental conditions
may have a somewhat pervasive influence, even though we may not always be
able to isolate their impact at some specific moment in time. Instead, their
impact may reflect the prevailing conditions confronting the individual over a
continuing period oftime. One example ofthis view ofthe environment in terms
of prevailing conditions, rather than as temporally discrete stimulus inputs, can
be found in adaptation-level theory (Helson, 1964).In this important work,
prevailing setting conditions are seen as establishing a perceptual-cognitive
frame of reference for the individuals on-going transactions with the environment. (For an excellent application of this approach, see Wohlwill and Kohn,
973.)
Second, and perhaps more commonly, the stream ofbehavior is redirected or
shunted by intentional acts on the part of the individual. In these cases, the
individual engages particular features of the environment in the course of some
activity. And significantly, through many of these acts, the individual can
modify the environment, e.g., by manipulating objects, by affecting others
around him, by constructing tools, or by building shelters. These environmental modifications, in turn, transform the surround, which may influence the
individual, and so on in a reciprocal manner (Shotter, 1983).As a result, the
individual may take an active role in fashioning his environment in important
ways. These on-going reciprocal exchanges form an essential part of the
individuals psychosocial history and, consequently, serve as the foundation
from which subsequent development proceeds. The reciprocal nature of social
interaction in particular has been receiving considerable attention, particularly
in the area of child development.
To summarize, in a relational view of environment-animal interaction, the
standard causal analysis may have only limited applicability. A mechanistic
10
Harry Hcfi
model typically imposes temporal discontinuity on environment-animal

transactions, and causal influences are usually seen as being unidirectional,
moving from environment to the animal. However, the nature of living systems
is such that there is an on-going transaction between an animal and its
surround, and environmental influences tend to constrain rather than elicit
behavior. Further, causal influences will be temporally continuous, reciprocal,
and cumulative.
What then is the causal character of the relationship between the affordances
of the environment and the perceiver? The relationship between an affordance
and behavior is that of fittedness and compatibility. While affordances are
features of the environment, they do not cause or elicit behavior, although
the presence of a particular affordance can sometimes prompt an act that
comports with it. More commonly, affordances constrain to a considerable
degree what actions may be expressed in a setting; or put in a positive way, they
create possibilities for particular activities. Affordances constitute the ecological
resources of the environment that may be utilized by the individual. Which
particular affordances are utilized in a given environmental setting will depend
on intentional processes of the perceiver. In advance of our examination of
intentionality below, we can state here that the causal relations between
affordance and perceiver are ( I ) continuous - rather than disjointed or
segmented; ( 2 ) reciprocal -in as much as typically affordances offer particular
possibilities for action, and perceptual functioning, by virtue of its intentional
character, realizes one (or more) of these possibilities; and (3) cumulative -in
that person-environment interactions at one point in time provide an historical
basis for subsequent perceptual development and environmental discovery.
We now turn then to a consideration of one particular aspect of the
perceiving process: the intentionality of perceiving. This characteristic of
perceiving is only implicitly expressed in Gibsons theory, but drawing out its
essential role in the ecologicalframework will allow us a more substantive grasp
of the affordance concept. In addition, this analysis will provide the basis for
extending the concept of affordance to the problem of perceiving culturallyderived meaning of environmental features.
AFFORDANCES AND INTENTIONALITY
Our own body is in the world as the heart is in the organism: it keeps the visible spectacle constantly
alive, it breathes life into it and sustains it inwardly, and with it forms a system. (Merleau-Ponty,
1963. P. 03)
As we have seen, the relational nature of affordances points to characteristics of

the individual, and in particular characteristics relating to an individuals body
scaling. Referring back to our previous examples, whether a surface affords
I1
sitting-on or an object affords grasping depends on some relevant body

dimension. Additional examples of body-scaled affordances include steps
(relative to leg length), doorways (relative to height and width), shelves
(relative to arm extension), and so on. These body-scaling criteria underscore
the individualistic, and thus the relational nature of affordances.
However, I would like to suggest that the affordances of the environment
refer to the body in a much more fundamental manner than merely bodyscalingper se. Affordances are specifiable relative to what an individual can do,
relative to what his potentialities for action are. That is, the environments
affordances are to be identified in relation to the body as a means of expressing
various goals or intentions.
This broader usage of body has been explored by Merleau-Ponty ( 1963),
whose phenomenological analysis of perception shares many features with
Gibsons ecological theory (see Glotzbach and Heft, rg82, for a preliminary
comparative discussion). For Merleau-Ponty:
The body is the vehicle for being in the world, and having a body is, for a living creature, to be
intervolved in a definite environment, to identify oneself with certain projects and be continually
committed to them. (Merleau-Ponty, 1963, p. 82)
In this statement, projects can be read as referring to specific goal-directed

actions of the person in the world, actions with a directionality and an end
implicit in their origins -in other word, intentional acts. Thus, Merleau-Ponty
is suggesting here that the body should be viewed more broadly than only in a
physical or physiological sense; it is the means by which goal-directed actions
that comprise the individuals intentional repertoire are expressed.
Merleau-Ponty goes to considerable lengths to emphasize that intentional
acts do not exist in the individual in the form of, e.g., a motor representation or
schema. Instead, he argues, intentional acts are always situated. That is,
inherent in an action is a reflection of a situation or a set of conditions. An
intention is not describable in the absence ofsome foreseeable expression ofit in
the world. In this respect, intention does not refer to a mental representation; it
is not a mentalistic notion. Rather, it refers to possibilities that are only
realizable as situated behavior.
To explore these ideas a bit further, an intentional act as a representation
independent of situational factors is at best an abstraction. Instead, intention
refers to possibilities that are only instantiated in a particular form in
interaction with situational factors (see below). In this respect, it is similar to the
concept of genotype. Here also we have a concept that reflects certain
possibilities that can be manifested in the presence of particular environmental
factors. The biochemical basis for genotype is very real, as is the functional
organization of the motor system. But what will be expressed phenotypically is
not preordained in the absence of particular environmental conditions, and
I2
Harv HeJ
likewise, specific motor actions (excluding perhaps certain simple reflexes) are
probably not programmed in the nervous system to be expressed without the
presence of situational factors that play a direct role in establishing their
particular form (Turvey, 1977).
An intentional act is situated with respect to two factors: the functional
characteristics of the environment confronting the individual, (i.e., its affordances), and the physical characteristics of the individuals body, (e.g., bodyscaling). The affordances of the setting are, in a sense, the ecological resources
for behavior. The physical characteristics of the body establish what can be
performed (i.e., what the individual can do) as a function of such things as
length of reach and stride, breadth of grasp, strength, etc. In combination, the
affordances of the environment and the characteristics of the body constrain the
range of intentional acts that can be expressed.
To take an example, sitting down is an intentional act with two situational
facets: perceiving a surface ofsupport that affords sitting and simultaneously an
awareness of ones relevant body dimensions. These two factors jointly create
the possibility for the expression of this intention and determine its particular
form. The act ofsitting down has little meaning as a pure motor representation;
it is only meaningful as a situated act.
Both Merleau-Ponty and Gibson emphasize that perceiving simultaneously
entails an awareness ofboth the environment and the body. Just as affordances
are perceived properties of the environment, our body is also phenomenally
present as we move around and engage the world. While the object or event
receiving our attention at a particular time stands out in relief as a figure
against the background of the rest of the environmental array,
ones own body is the third term, always tacitly understood, in the figure-background structure,
and every figure stands out against the double horizon of external and bodily space. (MerleauPonty, 1963; p. 101)
Gibson (1g7ga) makes a similar point about these two facets of perceptual
experience:
The optical information lo specify the self, including the head, body, arms and hands, accompanies
the optical information to specify the environment. The two sources ofinformation coexist . . ..The
supposedly separate realms of the subjective and objective are actually only poles of attention.
(p. I 16). . .. The continuous act of perceiving involves the coperceiving of the self. (p. 240)
It is suggested then that both the environment and the body are aspects ofthe
perceptual field, and moreover, that goal-directed actions are realized in
relation to these two factors. The body, in particular, is an instrument or tool
through which intentional acts that are directed toward environmental objects
are expressed. That is, the body needs to be viewed not merely as a set of
physical dimensions, but more deeply as a vehicle for being in the world.
Affordances reconsidered
Adopting the intentional approach to the body suggested in the foregoing
invites a modification of our earlier definition of affordances. Instead of solely
specifying an affordance relative to the size of some relevant body feature,
perhaps we should couch this specification in relation to the body as it
participates in a particular goal-directed act. That is to say, an affordance is
perceived in relation to some intentional act, not only in relation to the bodys
physical dimensions. So, for example, whether that object before me affords
grasping must be assessed relative to an intentional act (i.e., grasping), and not
only with respect to hand size, although this remains an essential factor.
This point may seem to entail a relatively minor definitional revision, but it
has significant consequences by broadening considerably the possibilities of
what can count as an affordance. If affordance is defined relative to the
intentions of the individual, instead ofonly relative to body-size considerations,
we can attribute functional meaning to any environmental feature that is
implicated in an intentional, goal-directed behavior.
Consider the following example: Can one justifiably say that a typewriter
affords typing? If affordances are limited to body-scaled objects, this claim
makes little sense and may seem to be an unreasonable application of the
affordance concept. Even though the design of a typewriter keyboard is scaled
to the hand, the act of typing goes beyond mere manipulation of keys. It is a
structured act both linguistically (in terms of language expression) and
motorically (in that a particular manipulation of the machines parts is
critical). At the same time, when viewed as a structured act, typing can be seen
as a goal-directed or intentional, situated behavior. The act of typing is realized
through the body in conjunction with a machine configured in a particular
way. Within the domain ofthis situated act, the typewriter affords a specific action,
namely, typing. From an intentional perspective, then, the typewriter takes on
functional meaning, that is, an affordance, within the context of this goaldirected activity.
We might then reconsider affordances as the functional significances of
environmental objects taken relative to what an individual can do with respect
to them. Knowing how to do something necessarily implicates( I ) the structural
characteristics of the objects utilized in the performance of the action and ( 2 )
the structural characteristics of the body that engages those objects. In other
words, knowing how to do something is situated knowledge.
Is the couching of affordances in an intentional framework consistent with
Gibsons theoretical approach? It clearly is appropriate to do so in that the
action-oriented definition of the concept of affordances suggests an intentional
perspective. In contrast to an emphasis solely on their body-scaled characteristics, an intentional view ofaffordances may be more in keeping with Gibsons
concern with the environments possibilities for action. Indeed, body-scalingper
se suggests little, if anything, about action.
Harry Hgt
Further, Gibsons continual emphasis on the perceiver as a seeker of stimulus

information testifies to his intentional leanings: [Perceiving] is the activity of
getting information from the ambient array of light (Gibson, 1g7ga, p. 147;
also see, Gibson, 1958; Gibson, 1966, Chap.2). However, it is not solely a matter
of the perceiver self-directing his exploration of the environment; as was stressed
earlier, properties of the environment, especially affordances, constrain and
control the discovery process. Gibson points out the reciprocal role that
intentional action and affordances play in the on-going process of keeping-intouch with the world.
And what about the intentionality ofperception when an observer is seeking information instead
ofsimply having it presented to him? . . .What sounds to me promising is to begin with the assumption
that active perception is controlled by a search for the affordances of the environment and that
active behaviour is controlled by perceiving those affordances. (Gibson, I 974, pp. 387-388)
More generally, there is a broad intentional quality to Gibsons analysis of

perceiving. Consider the following comments:
Perceiving is an achievement ofthe individual, not an appearance in the theater ofconsciousness. It
is a keeping-in-touch with the world, an experience ofthings rather than a havingofexperiences. It
involves awareness-of instead of just awareness. It may be awareness of something in the
environment or something in the observer or both at once, but there is no content of awareness
independent ofthat which one is aware . . . perception is not a mental act. Neither is it a bodily act.
Perceiving is a psychosomatic act, not of mind or of body but ofa living observer. (Gibson, 1g7ga,
P. 23F240)
This emphasis on perceptual awareness as always awareness-of places

Gibson squarely in line with intentional approaches to psychological phenomena, from Brentano and Act Psychology through most phenomenological
approaches. At the same time Gibson takes pains to point out that perceptual
awareness is not a mental phenomenon. He, like Merleau-Ponty, is seeking an
alternative way of discussing these problems that avoids, on the one hand, the
arid and value-free world of material processes, and on the other hand, the
insubstantial fiction of a mentalism that is not grounded in naturalism. I will
have more to say about this point in the concluding section of the paper.
Intentionalig and the locus 0faJordance.s

An intentional analysis of affordances also brings to the forefront the matter of
the locus of functional meaning. This analysis suggests that, strictly speaking,
the perceived affordant meaning of an object resides neither in the object,
considered independently ofan individual, nor in the mind of the beholder, but
that it emerges from their relationship.
Let me attempt to be clearer about this point, first by turning again to
Merleau-Pontys ( I 963) analysis of perception.
When we say that an animal exists, that it has a world, or that it belongs to a world, we do not mean
that i t has a perception or objective consciousness of that world. The situation . . . is not entirely
articulate and determinate. . . i t presents only a practical significance; it asks only for bodily
recognition; it is experienced as an open situation, and requires the animals movements, just
as the first notes of a melody require a certain kind of resolution . . . (p, 78)
By denying a n objective consciousness of the world, Merleau-Ponty is

denying that perceptual experience consists of the passive imposition on the
individual of an environment with fully determinate meaning. Rather,
perceptual experience entails the participation of the perceiver in bringing to
realization some of the potential meanings of environmental features. (See the
discussion of actual versus potential affordances below.)
[The intentional act] does not arise from objective stimuli but moves back toward them. and invests
them with a meaning which they do not possess taken singly as psychological agents, but only when
taken as a situation. I t causes them to exist as a situation, i t stands in a cognitive relation to them,
which means that it shows them up as that which it is destined to confront. (p. 79)
In other words, the perceived meaning of the situation arises from the
interaction ofthe environments functional possibilities and the intentions of the
individual, that is, from its place in a situated, intentional act. Its functional
significance is relationally determined.
This claim requires additional explication, and for this purpose let us
consider Deweys (1896) treatment of perceived meaning in his classic critique
of the reflex-arc concept in psychology. He argues in this analysis that the
perceived meaning of an environmental object emerges from a continuous,
transactional interchange between the individual and the environment. This
point becomes clearer when it is recognized that often the meaning or value of
an object changes with experience and with intention. Let us first consider the
case of the modification of object meaning as a function of experience. T o use
Deweys example, at one moment a candle flame may appear attractive to a
child, who accordingly reaches out to it, but at a later moment, after the child is
burned, the candle appears aversive. How can we understand this state of
affairs?
The initially perceived quality of attractiveness is not solely inherent in the
object (i.e., as a thing-in-itself), otherwise how could this quality change with
experience? Nor is it held subjectively by the child independent of the object;
the object cannot be construed in any way by the child. Its possible meanings are
constrained by the objects physical properties (e.g., its light and warmth).
Instead, the attractive quality arises out of the relationship between the object
and the child at a particular time. The naive child visually inspects the candle
(perhaps we might say, adopts an aesthetic attitude toward it) and in the
context of this intentional action, the candle appears attractive. This positive,
perceived quality encourages haptic exploration (i.e., it appears to afford
16
Harry H$t
touching). Initially at least, the candle appears attractive to touch, as well as to

look at, in relation to the intentional act of exploration.
Following a painful episode with the object, its perceived value as an-objectto-touch changes, or rather this optical-haptic dimension of the relationship
between the object and the perceiver becomes transformed. I n particular,
through the course of exploring the candles flame, the child discovers an
invariant, namely the invariant conjunction of its visual appearance and its
cutaneous warmth. Henceforth, the affordance of the candle is that which
burns when touched, even though it remains attractive in relation to visual
exploration. Dewey put it this way:
The burn is the original seeing, the original optical-ocular experience enlarged and transformed in
its value. I t is no longer mere seeing; it is seeing-of-a-light-that-means-pain-when-contact-occurs.
(Dewey, 1896, p. 3 5 ~ 3 6 0 )
Again, the perceived meaning of the object does not reside solely with the
object or the person, but it derives from the intentional relationship between the
two, which includes the history of their interaction. The quality of the initial
experience is reconstituted through the course of interaction with the object.
Viewing environment-behavior interaction as being continuous as well as
cumulative, as discussed above, the meanings of environmental features
naturally become transformed over time (Shotter, 1983).
The claim that the functional significances of environmental features emerge
from the relationship between the object and an individuals actions can
perhaps be more simply and clearly illustrated by the observation that the
perceived affordance of an object also changes as a function of intention. A
lighted candle not only affords pain, ifyou touch its flame, but more positively,
it also affords illuminating a dark place as well as heating a liquid such as water.
Which of these latter two dispositional qualities is realized in experience depends
on the individuals behavioral goals or intentions at a particular time. The
illuminating quality of a flame emerges from intentions concerning the lighting
of a setting; and its heating quality arises out of intentions such as the desire to
boil water. Both properties inhere in the object; a candle gives off both light and
heat. Which one of these properties is functionally significant and is experienced
at a particular time depends on the intentional character of the perceivers
actions at that time. (We will return to the issue of the multiple affordances of an
object .)
The apparently mentalistic tone of this intentional analysis may seem at odds
with Gibsons consistent criticism of mentalistic accounts of psychological
processes. If such a characterization of this analysis was appropriate, then
indeed it would be inconsistent with the thrust of Gibsons theory. But such a
reading is not necessary, and certainly is not desirable. As we have seen, an
intentional approach is generally consistent with Gibsons ecological view.
Further, a recognition of the role of the individuals actions in realizing the

functional meaning ofan object is compatible with the claim of the independent
existence of those affordances relative to the individual. However, rather than
take up this matter here, it is best deferred until after we consider how the
affordance concept might be applied to the problem of perceiving the
sociocultural-based meaning of objects.
Affordances and culturally-derived meaning of object3

Many of the examples of affordances that we have considered up to this point
are of a species-specific or transcultural nature; that is, they are applicable to
humans as a species, independent ofcultural influences. Affordances of this type
include surfaces of support, graspable objects, reachable surfaces, traversable
chasms, as well as such events as imminent impact ofan object moving toward a
perceiver, and the imminent collision of a moving perceiver with an obstacle
such as a vertical surface. One might grant that the affordance concept, and
more particuarly, the assertion that functional meaning is conveyed by stimulus
information, works well in cases such as these, where the functional significance
of an environmental feature is apt to be species-wide. However, much of the
functional meaning in our perceptual experience is not of this nature; it is
culturally-derived. And extending the concept of affordances to meanings that
are specific to a culture may seem unwarranted, or at the very least, a careless
and inconsistent application of the concept.
Can the affordance concept be applied to cases of culturally-derived
meaning, or is it to be limited only to those meanings ofa more transcultural or
species-specific nature? When affordances are viewed narrowly as body-scaled
features, in the manner discussed above, then indeed the concept will have little
to say about the culturally-derived meaning ofobjects. In this case, the concept
would have interesting, but somewhat limited usefulness in tackling the
problem of meaning in perception. However, if affordances are specified
relative to the individuals intentional repertoire, the prospects for a wider
applicability are promising.
The intentional acts that a person acquires within a sociocultural context,
like any goal-directed actions, are situated with respect to particular objects,
and those objects are invested with a functional meaning in relation to these
actions (as discussed in the previous subsection). For example, in this culture,
forks afford skewering food and transporting it to the mouth, pens afford writing
and drawing, mailboxes afford sending correspondence, telephones afford
verbally communicating at long distances, and so on. The functional meanings,
or affordances, of objects such as these reside within the relationship between a
specific object (the properties of which are specified by higher-order information in the ambient array) and a particular intentional act. The fact that
another person teaches the perceiver the behavioral significance of an
18
Harry Hejl
environmental feature, or that a feature has a socially-shared and conventional

meaning, does not make its significance any less real from a functional
standpoint. Ifaffordance is defined relative to the body in an intentional sense,
then an extension ofthe concept to this type offunctional meaning is warranted.
Let us consider this argument through an example. Gibson (1g7ga) has
suggested that a mailbox affords mailing a letter (cf. p. I 39). In what sense do
mailboxes have this functional meaning? They do to the degree that a person
knows what it means to mail a letter; that is, to the degree the act of mailing a
letter is a part of an individuals intentional repertoire. If this knowledge is
acquired in the United States, then essentially rectangular-shaped blue objects
with four legs and a curved top will perceptually mean a letter can be mailed
here. In Great Britain, a particular red cylindrically-shaped object will afford
the expression of this intentional act. In either case, there is information in the
ambient array specifying a container which is invariantly linked to the
intention mailing a letter. This intentional act is situated with respect to this
object, and its affordance emerges from the relationship, as designated by the
culture, between this particular object and this intention.3
This analysis suggests that the extension of affordances to the culturallybased meanings ofobjects is justified ifwe view affordances in relation to what
an individual can do, or rather what an individual knows how to do. Much of
this situated knowledge is acquired within a specific sociocultural context.
Moreover, the process of enculturation can be viewed, in part, as one of
acquiring a repertoire of acts, each act being situated with respect to a
particular set ofenvironmental features, the functional significance ofwhich are
socially conveyed.
Perceptual learning of afordances
How does an individual learn about environmental affordances whose

functional significances are culturally-derived? We might speculate that these
kinds of affordances are revealed by other members of the culture, particularly
parents, siblings, relatives, friends, and other valued sources of information. By
receiving direct instruction, or probably more typically through observing
others actions, the individual can learn the culturally-specific uses for
particular objects.
T o be more specific, the individual learns particular situated, intentional acts in
social contexts. In this process, there is simultaneously a patterning of motor
behaviors expressing a particular intention and (and in relation to) an
enhanced sensitivity to the information in the ambient array specifying the
related affordance. Developmentally, both facets of the affordance-behavior
relationship change over time. The individuals intentional repertoire expands
and differentiates, and concurrently, invariants specifying new affordances in
the environment are discovered. In the course of this kind of perceptual-motor
19
development, the individual learns about the affordances specific to a

particular culture, and this learning forms a significant part of the enculturation process more generally.
Ofcourse, one does not need others to convey the affordancesofobjects whose
functional significance is not culturally specific. In order to provide a more
complete account of affordance learning, let us briefly consider two other ways
in which intentional skills are differentiated and new affordances are revealed:
a . Perceptual learning that accompanies maturation and other sources of physical change
Some of the affordances of the environment are discovered as the individual
changes physically through maturation. Changes, for example, in height or
strength, permit new actions in the environment. The individual can perform
actions that were previously not possible. Conversely, in the case of aging or
significant changes in motor ability due to accident or disease, the individual
can no longer perform certain actions. The physical state of the body and motor
proficiency (and the accompanying awareness of these body factors) are one set
of factors that establish what one can do, that is, what intentional acts are
possible.
In turn, (as argued earlier), intentional possibilities invest the world with
functional meaning. For example, when the child develops the motor abilities to
grasp small objects and to make controlled arm movements, objects such as
pencils, pens, and crayons take on new functional significance; namely, they
afford marking, scribbling, drawing, etc. The discovery of new affordances is
made possible by the maturational change, and more deeply by enhancement
of the intentional repertoire.
Loss of function, or of intentional possibilities, through aging, disease, or
accident also results in changes in the possibility of realizing particular
affordances. For example, in the case ofsome disabilities, stairs may no longer
afford locomotion, but instead signify an obstacle. As the intentional capacities
of an individual change, the affordances of the environment change concurrently.
b. Perceptual learning through exploration

New affordances can also be discovered through exploration by the individual,
apart from maturational changes. In the course ofexploring objects and places
in the environment, the child fine tunes motor skills and simultaneously learns
more about environmental features. Put in another way, exploration and motor
skill refinement enhances the intentional repertoire and concurrently leads to
the uncovering of new environmental affordances. T o carry over the example
used above, a child with some minimal motor skills for drawing may discover
through personal exploration that lipstick tubes, like pencils and pens, also
20
H a r y Hgt
afford drawing. Likewise, an individual who experiences some loss of a

particular motor function may learn new affordances as the existing motor skills
become more highly developed and differentiated.
Age-related changes in the visual world

Whether through maturation or learning, or more precisely through their
interaction, age-related changes in the way in which the environment is
experienced have long been noted by developmental psychologists with a
phenomenological ~ r i e n t a t i o n .For
~ example, Werner ( 1948)describes the
childs world as a world of action, a behavioral sphere in which everything is
framed in terms of handiness and unhandiness, of efficaciousness and inefficaciousness (pp. 382-383), and the functional properties of this world
become transformed with age. H e offers as an anecdote the likely changes in the
world of action of a young boy who
at the ageofeight nolonger recognizes the sea which he knew at the age offour. At that time thesea
was determined by different things-of-action. Such small objects as mussels and little stones,
butterflies, and the wet sand ready to be molded into simple forms- these made up the world ofthe
seashore for the four-year-old, whereas the eight-year-old conceives this same region as an arena for
sports and swimming, and no doubt thinks of the tremendous surface ofthe water as an invitation to
adventure. (p. 383)
Among the other examples Werner offers is a summary of the pioneering,

empirical research of Muchow (Muchow and Muchow, 1935; also see
Wohlwill, 1985),who conducted one of the earliest observational studies of
childrens activities in a city. One site that was examined in the study was a
canal loading dock. Werner points out that observation at this site suggests how
different it must have been experienced by children as compared to adults.
For the adult the principal features ofthis place would be the street, the path down, and the landing
place. The child, particularly at an earlier school age, pays little attention to these elements. For
him the main features are the wooden fence and the slopes. T h e fence which, for the adult, has the
negative character ofstopping movement, is for the child, exactly to the contrary, the very signal of
movement. It invites the child to climb or jump o n it or over it. Similarly, the slopes, which would
have an indifferent or negative value for the adult, represent a provocative field of action for the
rhild. (pp. 38G387)
Putting this passage in the terms of the present discussion, it would seem that
the affordances of features at the loading dock site were different for adults
and children. Moreover, Muchow noted that the site was probably experienced
differently by children of the various ages observed. In particular, the fence
and the slope were perceived as affording different activities at successive age
levels:
Afordances and the Body
21
The child of two to three yean old frequently apprehends fence and slope as areas ofindependence,
as a means of affording a temporary, pleasurableseparation from the mother standing on the other
side of the rails. Children ofkindergarten age use the fence as something on which to sit and balance
themselves, whereas six- to eight-year-olds use i t as a gymnastic device. For the three- to four-yearold the slope is something to be apprehended in awe and curiosity as a field ofdanger. Somewhat
older children try to master this danger zone by sliding down it on their seats. (Werner, 1948,
PP. 387-3881
These phenomenological accounts illustrate some possible age-related changes

in the manner in which individuals experience aspects of the environment.
Further, they also serve as examples of the claims that afiordances of the
environment are relationally specified and independent properties of the
environment (see below), and that environmental features may have multiple
affordances. With regard to the latter, as the child discovers affordances of
environmental features, either through exploration, instruction, or modelling
others actions, she comes to recognize that objects can afford not only one
particular action, but also that they often have multiple affordance possibilities.
Let me conclude this section with a brief consideration of this matter and its
implications.
Multiple afordances of objects and the actual-potential distinction

Most objects can be used in many ways. A book is a common multi-purpose
object in the adult world. Not only is it a source ofinformation, but it can also be
used to prop open a window, to stop a door, to support a table, to decorate a
shelf, and so on. One of the delightful experiences of childhood is probably the
discovery ofnew uses for familiar objects, or put in other words, the discovery of
a new affordance in a familiar object. This type ofdiscovery was, in fact, used by
Kohler ( 1925) as one type of evidence for intelligence in primates in his famous
studies of insight learning.
The child can discover new affordances of an object by using the object in
new ways, as well as by observing others doing so. Spoons can not only be used
for eating, but also for digging in the dirt. Boxes can be used to store other
objects, but they can also be used as enclosures to hide in. Particularly popular
among children are those environmental features that have many possible use
functions, for example, sand and clay, as can be seen in Barker and Wrights
( 1 9 5 1 )observations of one boys activities during the course of a day (for
example, see pp. 338 ff.; also see Heft, in press).
The notion of the multiple affordance possibilities of an object is consistent
with the intentional analysis offered above. The affordances of an object are
realized in relation to some intentional act in the individuals behavior
repertoire. For example, ifan individuals goal or intention is to cut an object, a
second object that has a sharp edge and is graspable will be perceived as
affording cutting-with. If the individuals intention is to pry open a lidded
22
Harry Hdt
object, that same cutting-with object will be perceived as affording pryingwith. If the intention is to tighten a screw, this same object may afford that
particular action. In each of these cases, the various affordances of the object
arise from a relationship between a particular intentional goal and the
properties of an environmental object, as discussed at length above.
Typically, an environmental object offers a variety of potential affordances,
one (or more) of which is realized in conjunction with a particular intentional
act. The multiple affordance possibilities of an object highlight a crucial
distinction between potential and actualized afordances, a distinction that has its
counterpart in potential and actualized intentional acts. At any given time, the
individual is perceiving-utilizing only a subset of the potential affordances of an
object. In addition to the potential affordances of a single object, more generally
the individual is experiencing few of the affordance possibilities in the total
environment a t any particular time. The remaining environmental affordances
exist as possibilities to be realized in relation to the individuals intentions.
Following Turvey, Shaw, Reed, and Mace (1981),we can distinguish
between dispositional and occurrent properties of objects. Dispositional
properties refer to a things potentialities -to what can happen. As such they
are to be distinguished from occurrent properties, the properties that a thing is
currently exhibiting (p. 261). The dispositional properties of objects considered at a behavioral level are their functional possibilities-what an animal
can potentially do with them or in relation to them. These properties can be
specified as environmental potentialities, apart from their occurrence or
realization, but in relation to some animal (see below). Accordingly, functional
possibilities, or potential affordances, like any dispositional (e.g., the solubility
ofsodium chloride) ,are real properties ofobjects or substances. The ecological
approach. . . focuses on real possibility; for it takes possibility to be an
ontological category (Turvey, et al., 1981,p. 262).Thus, the environment for a
particular perceiver can be considered to consist of permanent possibilities
for perceptual-motor experience. We will pursue this issue further in the next
and concluding section.
ECOLOGICAL INDEPENDENCE AND THE RELATIONAL NATURE OF AFFORDANCES
As we have seen, there is a tension in Gibsons theory between two apparently

competing claims: On the one hand, the properties of the perceived environment, including affordances, are considered to be independent of the perceiver;
they constitute ecological reality. Perceiving is seen as a process of keeping in
touch with environmental conditions through the pick up of information in the
ambient array unequivocally specifying these conditions, as opposed to being a
ratiomorphic process of making inferences, hypotheses, or best guesses about
the environment from equivocal, sensory cues. Thus, Gibsons theory of
23
perception provides possible grounds for the epistemological position of direct

realism as an alternative to the indirect realism assumed by the standard
constructivist approaches to perception (Gibson, I 967).
O n the other hand, Gibson advocates a relational approach to perception, as is
most clearly reflected in the affordance concept. The environments affordances
are specified relative to an individual, or put in another way, attributes of the
perceiver contribute to the specification of psychologically significant environmental properties. This approach would suggest that environmental
features, and affordances in particular, do not exist independently of a
perceiver.
Is there a fundamental contradiction between these respective claims that
environmental features, and affordances in particular, exist independently and
that they are relational in nature? If not, what is the basis for this apparent
contradiction, and how can it be resolved?
Noble ( I 98 I ) has cogently discussed this dilemma, and he sees the problem
arising from Gibsons failure to acknowledge and to develop sufficiently the
pragmatist tendencies in his own theory.
Gibsons whole theoretical program has derived much ofits power from insistence on the organism
as an active investigator ofits environment . . ..But his distaste for idealism makes him overlook the
pragmatic nature of the act . . . in the production of an organisms perceptual world. He can see
that organismic activity is vital in any valid account ofperceptual experience, but he cant see that
the act itself- what the organism is doing, is infmding to do, can do and cant d o -is . . . bound up
in the generation ofits perceptually experienced world. (p. 70)
Whereas Gibson stresses the critical role played by the perceivers actions in
the perceptual process, specifically in the pick up of invariants in the ambient
array, he is wary to promote more fully the pragmatic or intentional role of
action in perceptual experience. The reasons for this are all too obvious. Gibson
wants to avoid the appearance of introducing mentalistic concepts into his
account of perceiving so that he cannot be read as advocating a mentalistic or
dualistic approach, or an enrichment theory (Gibson and Gibson, 1955).
Accordingly, he seems to limit his discussion of the perceivers role in the
perceiving process, particularly (but not solely) through the language that he
employs.
Noble feels that Gibson has unjustifiably drawn away from the pragmatic
and intentional character of his own position, and
that Gibsons neglect of the pragmatist tradition. . . has meant a lack of access to nonmcntulistic
concepts ofthe organisms role in the creation ofits experience. Notions such as intentionality, for
apsychologisf, belong to motivation, and for Gibson, I suspect, that would be classed as part ofthe
baggage for mentalism and put to one side. (pp. 7-71; emphasis added)
Noble contends, and I concur, that one can adopt an intentional approach to
perceiving, and with it acknowledge more fully the individuals participation in
24
Harry H g t
the perceiving process, without necessarily falling into the camp of mentalism
(also see Ben-Zeev, 1984).Gibson could have articulated more explicitly this
facet of his theory without being overly concerned about this latter possibility
(but not the possibility of being misinterpreted). Indeed, in the preceding
section of this paper I tried to develop the intentionality in Gibsons theory,
while attempting to remain faithful to the nonrnentalistic commitment of his
position. Moreover, it is only by explicating the intentional strain in Gibsons
ecological theory that the apparent tension can be resolved between the
independence and the relational claims concerning affordances.
Before demonstrating why this contradiction is more apparent than real, let
us first consider this intentional approach more broadly. If one is inclined to
assimilate all psychological concepts to a dualistic framework, then intentionality will seem to refer to a mental process. This assimilative tendency toward
dualism is commonplace in perceptual analyses; indeed Gibsons theory has
often been misread or failed to be understood because it does not fit into a
dualistic framework (Heft, I 980, I 982; Wilcox and Katz, I 98 I ). But intentional
approaches to perception are typically viewed by their proponents as offering
an alternative to dualism, and more particularly, as an alternative to the two
manifestations of dualistic thinking in psychological theory: behaviorism and
mentalism. Commenting on Merleau-Pontys intentional analysis of behavior,
Wild (1963)states:
Human behavior is neither a series of blind reactions to external stimuli, nor the projection of
acts which are motivated by the pure ideas of a disembodied, worldless mind. It is neither
exclusively subjective nor exclusively objective, but a dialetical interchange between man and
world, which cannot be adequately expressed in traditional causal terms. . .. It is out of this
dialectical interchange that human meanings emerge. These meanings are neither passively
assimilated from an external, cosmic order that is already fixed and established, as the realists have
imagined nor constructed dc nouo by a creative mind, as the idealists have supposed. (pp. xiv-xv)
Similarly, as we saw above, an ecological approach to perception rejects both a

physicalistic-mechanisticaccount of animal-environment relations and a mentalistic account that is not grounded in naturalism. Instead, what is needed is a
single theoretical language - in the spirit of animal-environment synergy that manages to incorporate both the objectivity of the physical language and
the agent orientation of the phenomenal language (Shaw and Turvey, 1981,
p. 365). With the concept of affordance Gibson sees himself as offering the kind
of alternative theoretical language referred to in the preceding passage.
An affordance, as I have said, points two ways, to the environment and to the observer. So does the
information to specify an affordance. But this does not in the least imply separate realms of
consciousness and matter, a psychophysical dualism . . .. This is only to re-emphasize that . . . to
perceive the world is to coperceive oneself. This is wholly inconsistent with dualism, in any form,
either mind-matter dualism or mind-body dualism. T h e awareness of the world and ones
complementary relations to the world are not separable. (Gibson, 1979a, p. 141)
25
And
But, actually, an affordance is neither an objective property nor a subjective property; or it is both if
you like. An affordance cuts across the dichotomy ofsubjective-objective and helps us to undentand
its inadequacy. It is equally a fact ofthe environment and a fact ofbehavior. I t is both physical and
psychical, yet neither. (Gibson, 1g7ga, p. 29)
How then does this alternative approach, reflected in the affordance concept,
allow us to reconcile the tension between the independence and relational
claims? In the ecological approach, the environment for an individual is
delimited relationally in terms of what is potentially significant, in a functional
or an intentional sense, for that individual, such as what the potential
affordance properties of the environment are for a particular individual. As
suggested above, a n emphasis on the individuals potential for activity, i.e., the
character and range of her intentional repertoire, specifies particular environmental features with respect to which these actions are situated. Although
these environmental features are relationally specified, they can be considered
to exist independently of the perceiver in at least two (related) respects.
First, as argued above, affordances can be viewed as dispositional properties
of environmental features taken at a behavioral level of analysis; and viewed in
this manner, affordances can be appropriately considered to inhere in
environmental features. That is, the affordance properties that an object can
take on are constrained by the physical characteristicsofthe object. Affordances
are not imposed on objects by perceivers mental processes. Second, and
importantly, as a property of an object, an affordance is carried by the
information in the ambient array. Thus, an affordance like any other object
property specified by information in the ambient array, is always there to be
perceived.6 For these reasons, Gibson (1979a) is justified in stating:
The affordance ofsomething does nof change as the need of the observer changes. The observer may
or may not perceive or attend 10 the affordance, according to his needs, but the affordance, being
invariant, is always there to be perceived. An affordance is not bestowed upon an object by a need of
an observer and his act of perceiving it. The object offers what i t does because it is the object it is.
(PP. 138-1391
We can describe, then, a domain of relationally specified affordances, and

within that domain, these affordances can be properly said to exist independently ofa perceiver as potential functional properties of the environment (also see
Ben-Zeev, I 984).
Further, as discussed previously, in the course of the individuals on-going
activity, particular affordances will be experienced (i.e., actualized) in
conjunction with particular intentional actions; these affordances both complement and constrain these intentional processes. That is, the expression of
intentionality will bring to psychological realization, at a particular time,
26
Harry Heft
certain affordances that are compatible with these intentional acts. Needs
control the perception ofaffordances (selectiveattention) and also initiate acts
(Gibson, 1975, p. 411).Thus, among the affordance possibilities of the
environment, which are ontologically real, some affordances will be realized in
the course of the individuals interaction with the environment.
As the foregoing considerations indicate, by bringing out the intentional
quality of Gibsons theory, the apparent contradiction between the independent and relational nature of affordances is resolvable. An emphasis on the
pragmatic and intentional character of perceiving allows one to claim that the
functional significances of objects are to be specified relative to an individual
perceiver, while at the same time preserving the independent status of object
meaning (i.e., direct realism). Further, it can be seen from the foregoing
analysis that this tension grows out of a tendency to view perceptual processes
from a dualistic perspective, in which relational concepts such as affordances,
are inappropriately viewed as either objective, and independent of perceivers, OY
subjective, and dependent on perceivers.
CONCLUDING COMMENTS
I hope that this discussion of affordances has promoted a deeper understanding

of the concept by detailing its explicit features and by drawing out some of its
implicit theoretical roots. Further, by developing the latter, particularly the
intentional character of Gibsons ecological theory, I hope to have made a
convincing case for the possibility of extending affordances to instances of
culturally-derived functional meaning. In all of this, I tried to remain consistent
with the spirit of Gibsons intentions for his theory, as I have come to
understand them.
Of course, many of these theoretical arguments demand empirical validation. Some support for the affordance concept has appeared already, but
considerably more research is required to lend weight to its validity. Apart from
its ultimate validity, if consideration of this concept leads researchers to
investigate the meaningful dimensions of perceptual experience -a heretofore
largely neglected topic - then the concept of affordance will have made an
important contribution to the study of perception.
Harry H e j
Department of Psychology
Denison University
,Granville
OH43023
Afordances and the Body
27
Acknowledgements This paper is dedicated to the memory of Joachim F.

Wohlwill, who contributed significantly to the study of the relationship
between the environment and the individual.
I would like to thank the following individuals for the helpful comments
offered on an earlier draft of this paper: Robert Daubert, Philip Glotzbach,
Steen Halling, Claus Seligman, and Bahram Tavakolian. Work on this
manuscript was partially supported by a Robert C. Good Fellowship awarded
through Denison University. The project was completed while the author was a
Visiting Scholar in the Department of Psychology and in the College of
Architecture and Urban Planning, University of Washington, Seattle,
Washington. I am most grateful for the supportive environment and resources
provided through the auspices of the Center for Planning and Design, David
Bonsteel, Director, and the Department of Architecture, College of Architecture and Urban Planning, University of Washington.
NOTES
To simplify the discussion, I will focus primarily on vision throughout the paper.
Turvey and Shaw (1979;also see Shaw and Turvey, 1981)discuss the compatibility
relations between environment and animal in terms of affordances and their complement, on the animal side of the ledger, eJectivities. Their emphasis on an animals
potential for action and its relationship to affordances is broadly similar to the
intentional analysis presented in this paper. Although the Shaw-Turvey approach
influenced my thinking in a general manner, I chose to develop the present form
of intentional analysis, rather than follow their lead for several reasons; most
notably because this intentional approach seemed more readily to offer an avenue to the
problem of perceiving the culturally-based meaning of environmental features (see
below).
This perceptual knowledge does not include a n understanding of how the postal
system works (Noble, 1981).Such understanding is of an abstract, cognitive nature in
that it entails at least a rudimentary knowledge of the workings of this institution. One
can know that a particular environmental object affords the mailing of correspondence,
without an understanding of what happens after the envelope disappears down the
chute.
* For a preliminary discussion of the role of affordances in the development of
environmental knowing in children, see Heft and Wohlwill (1987).
See Noble (198I ) for a suggested resolution of this contradiction by drawing on the
work of G. H. Mead.
Shotter (1983)is critical of Gibsons theory on similar grounds as Noble (198I ) ,
arguing that Gibsons account lacks an intentional analysis ofthe perceivers activities in
perceiving processes. Toward this end Shotter presents a stimulating application of
Giddens ( I 979) concept of structuration to individual-environment transactions
(whereas Giddens is primarily concerned with social structures). While I agree that the
intentional aspects of perceiving need to be articulated in Gibsons ecological theory, I
contend (as I believe Noble does) that this aspect of perceiving is already implicit in
Gibsons account (see above and subsequent comments in this section). Shotter fails to
recognize this.
28
Harry HeJ
Moreover, Shotter misattributes a hidden Platonism (of all things!) to Gibsons

theory. According to Shotter, Gibson views perception as a two-step process: first the
animal perceives an affordance, and second the animal decides how to make use of it
(p. 28). Shotter offers no evidence for the validity ofthis account ofGibsons theory, and
his justification for this strange rendering is unclear to me.
It is important to reiterate the critical distinction in Gibsons account between
potential information in the ambient array, which is available to be picked up, and
information that is picked up at a particular time; and thus between potential
affordances and affordances that are perceived at a particular time.
Practically speaking, a description of the environment in terms ofthe potential actions
that can be taken within it is a product of the architectural design process. Ideally, at
least, architectural design provides functionally suitable places for a particular user
group (Le., users with particular functional needs). Functionally successful designs
require anticipation of what an environment affords an individual prior to any action
taken within it, indeed independent of whether anyone is there in the environment to
perceive its affordances or not.
REFERENCES
BARKER,R.G.
(I
968). Ecological psycholog. Stanford, CA: Stanford University
Press.
BARKER,
R.G., and WRIGHT,H.F. (1951).One boys day. New York: Harper and
Row.
BARKER,R.G., and WRIGHT,H.F. (1955).Midwest and its children. New York:

Harper and Row.
BEN-ZEEV,A. (1984).T h e Kantian revolution in perception. Journal f o r the

Theory of Social Behavior, 14,6984.
DEWEY,
J. ( I 896). T h e reflex arc concept in psychology. Psychological Review, 3,
357-370.
DEWEY,
J. (1934).Art as experience. New York: G.P. Putnams Sons.
DEWEY,
J. and BENTLEY,
A. ( I 949).Knowing and the known. Boston: Beacon Press.
GIBSON,
E.J. (1969).Principles of perceptual learning and development. New York:
Appleton-Cen tury-Crofts.
GIBSON,E.J. (1982). T h e concept of affordances in development: T h e

renascence of functionalism. I n W.A. Collins (ed.), The concept of development.
The Minnesota symposium on child development. Vol. I 5. Hillsdale, NJ: Erlbaum
(PP. 55-8 1 ) *
GIBSON,
E.J., a n d SPELKE,
E. (1983).T h e development of perception. I n P.H.
Mussen (ed.), Handbook of childpsychology (4th Ed.) Vol. 3. New York: Wiley.
(PP- 1-76).
GIBSON,
J.J. (1958).Visually controlled locomotion and visual orientation in
animals. British Journal of Psychology, 49, I 82- I 94.
GIBSON,
J.J. ( 1966).The senses considered asperceptual systems. Boston: HoughtonMifflin.
GIBSON,
J.J. (1967).New reasons for realism. Synthese, 17, 162-1 72.
Ajordances and the Body
29
GIBSON,
J.J. (1972).The affordances of the environment (previously unpublished manuscript). In E. Reed and R. Jones (eds.), Reasons f o r realism:
Selected essays ofJames 3. Gibson. Hillsdale, NJ: Erlbaum (pp. 408-410).
GIBSON,
J.J. (1974).The theory of proprioception and its relation to volition:
An attempt at classification (previously unpublished manuscript). In E.
Reed and R. Jones (eds.), Reasonsf o r realism: Selected essays o f James 3. Gibson.
Hillsdale, NJ: Erlbaum (pp. 385-392).
GIBSON,J.J. ( 1975). Affordances and behavior (previously unpublished
manuscript). In E. Reed and R. Jones (eds.), Reasonsf o r realism: Selectedessays
ofJames J. Gibson. Hillsdale, NJ: Erlbaum (pp. 4 1 e 4I I ).
GIBSON,J.J. (1979a). The ecological approach to visual perception. Boston:
Houghton-Mifflin.
GIBSON,
J.J. (1979b). What is involved in surface perception? (previously
unpublished manuscript). In E. Reed and R. Jones (eds.), Reasonsf o r realism:
Selected essays ofJames J . Gibson. Hillsdale, NJ: Erlbaum (pp. 1 0 6 - 1 I 2 ) .
GIBSON,
J.J., and GIBSON,
E.J. ( 1955).Perceptual learning: Differentiation or
enrichment? Psychological Review, 62, 32-4 I .
GIDDENS,
A. ( I 979). Central problem in social theory. London: Macmillan.
GLOTZBACH,
P.A. and HEFT, H. (1982).Ecological and phenomenological
approaches to perception. Nous, 16, 108-1 2 I .
GRENE,
M. ( I 963). A portrait ofdristotle. Chicago: University of Chicago Press.
HALLFORD,
W. ( I 984). Sizing-up the world: The body as a referent in a sizejudgment task. Doctoral dissertation. Ohio State University, Columbus,
OH.
HEFT,H. ( 1980).What Heil is missing in Gibson. Journal f o r the Theory of Social
Behavior, 10 187-193.
HEFT,H. ( 1982).Incommensurability and the omission in Gibsons theory.
Journal f o r the Theory of Social Behavior, 12, 345-347.
HEFT, H. (in press). Affordances of childrens environments: A functional
approach to environmental description. Childrens Environments Quarterly.
HEFT,H. and WOHLWILL,
J.F. (1987).Environmental cognition in children. In
D. Stokols and I. Altman (eds.), Handbook of environmental psychology. New
York: John Wiley (pp. 175-203).
HELSON,
H. Adaptation-level theory. New York: Harper, I 964.
KOHLER,W. ( 1925). The mentality of apes. New York: Harcourt, Brace and
World.
MARK,L.S. ( I 987).Eyeheight-scaled information about affordances: A study of
sitting and stair climbing. Journal of Experimental Psychology: Human Perception
and Performance, 13,3617370.
MERLEAU-PONTY,
M. ( I 963). The phenomenology ofperception. ( C .Smith, Trans.)
London: Routledge and Kegan Paul. (Original work published I 945).
MUCHOW,
M. and MUCHOW,
H.H. (1935).D e r Lebensraum des Grosstadtkindes.
Hamburg: M. Riegel.
30
Harry Heft
NEISSER,
U. (1976). Cognition and reality. San Francisco: Freeman.
NOBLE,
W.G. ( 198I ) . Gibsonian theory and the pragmatist perspective. Journal
f o r the Theory of Social Behavior, I I , 65-85.
REED,E., andJONES, R. ( I 981). Is perception blind?: A reply to Heil. Journal for
the Theory of Social Behavior, I I , 87-9 I .
REED,E. andJoNEs, R. (eds.) (1982).Reasonsforrealism: SelectedessaysofJamesJ.
Gibson. Hillsdale, NJ: Erlbaum.
SCHIFF,W. (1965).The perception of impending collision: A study of visually
directed avoidant behavior. Psychological Monographs. 79, (Whole no. 604).
SHAW,R.S. and TURVEY,
M.T. (1981).Coalitions as models for ecosystems: A
realist perspective on perceptual organization. In M. Kubovy and J.R.
Pomerantz (eds.), Perceptual organization. Hillsdale, NJ: Erlbaum (pp. 343415).
SHOTTER,J. (1983). Duality of structure and intentionality in an
ecological psychology. Journal f o r the Theory of Social Behavior, 13, 19-43.
TURVEY,
M.T. (1977). Preliminaries to a theory of action with reference to
vision. In R.S. Shaw and J. Bransford (eds.), Perceiving, acting, and knowing.
Hillsdale, NJ: Erlbaum (pp. 2 I 1-265).
TURVEY,
M.T. and SHAW,R.S. (1979). The primacy of perceiving: An
ecological reformulation of perception for understanding memory. In L.G.
Nilsson (ed.), Perspectives on memory research. Hillsdale, NJ: Erlbaum.
TURVEY,
M.T., SHAW,R.S., REED,E.S. and MACE,W.M. (1981).Ecological
laws of perceiving and acting: In reply to Fodor and Pylyshyn. Cognition, 9,
237-304.
WACHS,T.D. and GRUEN,G.D. ( I 982). Early experience and human development.
New York: Plenum Press.
WARREN,
W.H. (1984). Perceiving affordances: Visual guidance of stair
climbing. Journal of Experimental Psychology: Human Performance and Perception,
10,683-703.
WARREN,
W.H., and WHANG,S. (1987).Visual guidance of walking through
apertures: Body-scaled information for affordances. Journal of Experimental
Psychology: Human Perception and Performance, 13, 37 1-383.
WERNER,
H. ( I 948). Comparative psychology of mental development. New York:
International Universities Press.
WILCOX,
S. and KATZ,S. ( 1981).What Gibson isnt missing after all: A reply to
Heil. Journal f o r the Theory of Social Behavior, I I , 3 I 3-3 I 7.
WILD,J. ( 1963). Foreword to M. Merleau-Ponty, The structure of behavior (A.L.
Fisher, Trans.). Boston: Beacon Press. (Original work published 1942).
WOHLWILL,
J.F. ( I 985). Martha Muchow and the life-space of the urban child.
Human Development, 28, 200-209.
WOHLWILL,
J.F. and KOHN,I. ( 1973). The environment as experienced by the
migrant: An adaptation-level view. Representative Research in Social Psychology,
4,135-164.

Affordance Gibson PDF

Uploaded by

Copyright:

Available Formats

Affordance Gibson PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Affordance Gibson PDF

Uploaded by

Copyright:

Available Formats

Journal for the Theory of Social Behaviour: r g : ~

Affordances and the Body:

AN ECOLOGICAL APPROACH TO PERCEIVED MEANING

If we adopt a phenomenological attitude to perceiving, it is apparent that

How do we account for this dimension of perceptual experience?

According to the standard analysis of perception, meaning is imposed on

Affordances and the Body

as a perceiver stands at the edge of a cliff, the slightest movement results

However, affordances deviate from a strict meaning of an objective property

Affordances and the Body

express the functional compatibility existing between an animal and its

Causality, environment-animal relations, and affordances

Affordances and the Body

relationship between environment and animal is not substantively different

In fact, only about one-third of Mauds actions corresponded to immediate

Affordancesand the Boaj

model typically imposes temporal discontinuity on environment-animal

AFFORDANCES AND INTENTIONALITY

As we have seen, the relational nature of affordances points to characteristics of

Affordances and the Body

sitting-on or an object affords grasping depends on some relevant body

In this statement, projects can be read as referring to specific goal-directed

Affordances and the Body

Further, Gibsons continual emphasis on the perceiver as a seeker of stimulus

More generally, there is a broad intentional quality to Gibsons analysis of

This emphasis on perceptual awareness as always awareness-of places

Intentionalig and the locus 0faJordance.s

Affordances and the Body

By denying a n objective consciousness of the world, Merleau-Ponty is

touching). Initially at least, the candle appears attractive to touch, as well as to

Affordances and the Body

Further, a recognition of the role of the individuals actions in realizing the

Affordances and culturally-derived meaning of object3

environmental feature, or that a feature has a socially-shared and conventional

How does an individual learn about environmental affordances whose

Affordances and the Body

development, the individual learns about the affordances specific to a

b. Perceptual learning through exploration

afford drawing. Likewise, an individual who experiences some loss of a

Age-related changes in the visual world

Among the other examples Werner offers is a summary of the pioneering,

Afordances and the Body

These phenomenological accounts illustrate some possible age-related changes

Multiple afordances of objects and the actual-potential distinction

ECOLOGICAL INDEPENDENCE AND THE RELATIONAL NATURE OF AFFORDANCES

As we have seen, there is a tension in Gibsons theory between two apparently

Affordances and the Body

perception provides possible grounds for the epistemological position of direct

Similarly, as we saw above, an ecological approach to perception rejects both a

Affordances and the Body

We can describe, then, a domain of relationally specified affordances, and

I hope that this discussion of affordances has promoted a deeper understanding

Afordances and the Body

Acknowledgements This paper is dedicated to the memory of Joachim F.

Moreover, Shotter misattributes a hidden Platonism (of all things!) to Gibsons

968). Ecological psycholog. Stanford, CA: Stanford University

BARKER,R.G., and WRIGHT,H.F. (1955).Midwest and its children. New York:

BEN-ZEEV,A. (1984).T h e Kantian revolution in perception. Journal f o r the

GIBSON,E.J. (1982). T h e concept of affordances in development: T h e