This is an addendum to an article that I previously posted on affordances in April, 2016. As I’ve continued to explore the topic of affordances, there are a couple of things in the original article that I’ve reconsidered and need to address. The first is regarding some incorrect statements that I made about the persistence of affordances. The second is that I don’t think that Norman’s version of the concept of affordances needs to be “fixed” and brought in line with Gibson’s thinking. Rather, I choose to see it as a distinct concept that, confusingly, bears the same name as Gibson’s.
The persistence of affordances.
In the section “Norman’s affordances” in my original article, I suggest that when an observer does not perceive an affordance, then there is no affordance for that individual. According to my current understanding of Gibson’s theory of affordances, this is incorrect. Affordances are “invariant”, i.e. they always exist whether an individual perceives them or not. What differs between individuals is the meaning that the observed object takes on for a specific observer. The meaning is derived from the affordances that an individual’s attention is directed toward.
In the original article I overlooked the role of meaning in Gibson’s account of the perception process. Yet, it is perhaps the most important for making sense of the differences between Gibson’s and Norman’s accounts of affordances. The purpose of the theory of affordances is to account for how objects perceived in an environment become meaningful to an observer. For example, how does a shoe come to mean “object-for-protecting-one’s-feet” or (and perhaps at the same time), “object-for-squashing-bugs”? For Gibson, meaning emerges when an observer’s attention is directed toward an affordance that corresponds with an action that she wishes to perform. I’ve illustrated this in the figure below:
For Norman, the process is different. For Norman, meaning precedes the affordance. This is a necessary consequence of Norman’s dualistic position (or indirect perception). Meaning is a mental phenomenon (he refers specifically to “mental models”) that is brought to bear on the physical environment to reveal affordances. Norman’s process then looks something like this:
So, I am incorrect when I say in the original article:
“Otherwise, the object simply does not afford the action that I want to perform, i.e. there is no affordance.”
In this sentence, I am, in fact, not talking about affordances, but rather meaning. If the object ever can afford a given action, it always affords that action. But, although an object affords an action, the object will not necessarily come to mean something that corresponds with that action in every environment.
So, basically, the gist of this is that, for Gibson, meaning comes and goes while affordances are forever.
I came to this realisation while reading Shaleph O’Neill’s excellent chapter on the theory of affordances in his Interactive Media: The Semiotics of Embodied Interaction (see his comments on McGrenere & Ho on pg. 55). O’Neill and I would seem to be in agreement on a number of things, but we both made the same mistake regarding the persistence of affordances.
The other thing that I want to comment on is that I no longer regard Norman’s conceptualisation of affordances as a mistake in need of fixing (as O’Neill does). Although Norman’s version of affordances probably originates out of some misunderstanding of Gibson’s theory, it has taken on a life of its own and has proven useful for many things. The problem is that Norman’s affordances are not Gibson’s affordances, yet the two confusingly go by the same name. It is that we have two distinct concepts, both of which would seem to have a right to their existence as long as they are applied appropriately, that are both referred to as affordances that is confusing. What needs to happen (and I would pass this project along to others) is to clarify what Norman’s affordances are if they are not affordances in the Gibsonian sense, and perhaps advocate for a renaming.