One way of distinguishing between a cultural good and a medium is to say that a cultural good mediates meaning between a community and an individual, whereas a medium tends to function between an individual and another individual. However, this too quickly breaks down when it comes an environment’s mediating power. For example, if I decide to take my wife on a date, the experience of that data will be different depending on whether I take her to a nice restaurant or or a place that doesn’t pass inspection. The restaurant is not literally between her and me, but when I choose one over the other it still mediates something from me to her. Likewise, if I take a shower and dress well, the outcome is likely to be different than if I show up a stinky mess.
I'm not convinced it's helpful to think of environments as mediation, if only because expanding the content away from something that stands between two people inevitably waters it down. We can make our conceptual analysis too narrow or too broad, and I think John has fallen prey to the latter. That a particular environment alters our behavior and makes certain responses more plausible does not entail that we should treat it as mediation in the way a screen mediates our interaction.
One interesting aspect of the restaurant example is that the intentionality is built into the case. John chooses a particular restaurant, presumably for a specific set of reasons. Hence, it mediates content to his patient wife (whom we hope manages to get him to discuss something other than the topic at hand on the date). But in what way does the environment "mediate" his experience with the other patrons, or the waiter? Remove the degree of intentional decision beforehand, and the non-mediatorial aspect of the environment becomes, I think, a little clearer.
One other worry: In his section on language as mediation, John writes:
The power that resides in a name can be found all the way back to the Garden before the Fall when God asked Adam to name the animals. We know this was to find Adam a helper (Gen 2:18), but God also seems interested to find out “see what [Adam] would call them” (Gen. 2:19). Imagine that Adam sees a bird and decides to call it a “bluejay.” No big deal right? Well, it actually is a big deal, because he has just determined that the defining characteristic of that bird is how it looks. If the next bird comes along and he names it “woodpecker” he would have defined the bird according to its action, not its appearance.
These names are not mere neutral tools, objectively transferring bits of information between data terminals. Rather, names, words, and all language, sit between us and a thing, and language mediates certain aspects of the thing to us, shaping what we see and don’t see about a thing.
Not to go all Saul Kripke on you, but John has deployed the easiest examples for his case. What if it was a name like "Matthew Lee Anderson?" John's examples are naming things according to function or form, but it's difficult to see what sort of content is being mediated in other proper names. Or even names of colors. We might say that language "mediates" the content of redness to us, but that doesn't prevent us from having a real encounter with the color prior to learning the name.
A simple rule, perhaps: if everything is mediation, then nothing is.
Matthew Lee Anderson is the Founder and Lead Writer of Mere Orthodoxy. He is the author of Earthen Vessels: Why Our Bodies Matter to our Faith and The End of Our Exploring: A Book about Questioning and the Confidence of Faith. Follow him on Twitter or on Facebook.