Play involves straying from the expected, using the familiar in unfamiliar ways, overturning expectations, and pushing boundaries. At least in part, that means that something in a conversation or interaction has to nudge the thought processes of the individuals in question "off the beaten path," ultimately leading the conversation or interaction there as well. Part of what makes play so satisfying is the undercurrent of surprise that ties it to everything else and the "delicious sense of security" (as I imagine C.S. Lewis might phrase it) that the relationship will "resume its original shape" (like a rubber band) once the boundary is no longer being stretched and the tension has been released.
I think teasing is used (subconsciously) on a variety of levels—to map the boundaries of a friendship, to cement a friendship, to affirm a friendship, and to deepen/expand a friendship. You've probably noticed that as people become more comfortable with a friendship or a relationship of any kind, they begin to recognize the natural ebb and flow of the relationship (how the conversation "feels") without having to rely on external signposts, affirmations, or clarifications (such as the phrase, "Just kidding!", facial expressions, or a punch in the arm that affirms that one is still accepted by the person or group in question). In other words play is so much fun not only because of the fun inherent in the activity itself, but also because of the way it feels and what it represents/symbolizes about the parties involved.
Surprise is difficult to achieve without having first established a baseline of expected or "normal" behavior. I have observed that people often use verbal teasing or physical "banter" as a sign of belonging to a particular group. If you are "in," you can get away with it. If you are "out" and you try to replicate the playful behaviors that you have observed, your reception will be hostile at worst, cold at best. Why? Because although you can imitate the behaviors, you cannot imitate the current of energy responsible for the natural ebb and flow behind them (you can't be sure whether they are using 120 or a different voltage, to further the electrical analogy), or the foundation of trust that allows that exchange of energy to occur (wiring, size of the line). It is hard to play tug of war (representative of the ebb and flow of play) until both parties take hold of the rope (a.k.a. friendship/relationship of some sort) AND until both parties accept that inherent in play is the state of being off-balance, the willingness to be surprised.
Because play is ultimately based (at least in part) on the manipulation (or total breakage) of "the rules" (in terms of communication, content, culture, community, and the relationship in question—invisible though they may be), it also generally requires a deep (though, perhaps, unconscious) understanding and internalization of those rules and their accompanying conventions, allusions, and cultures (both geographical and the culture of a particular person or group), among other things. In other words, the most satisfying forms of play are based on understanding and on trust (partially involving the belief that equilibrium will eventually be restored and that the cost of making oneself vulnerable enough to play will be worth it). Picasso is a great example of this. He certainly is famous for breaking the rules, but what distinguishes him as a master artist is that he understood the rules implicitly when he set out to break them—thus allowing him to control (or, at least, to predict) the outcomes.
This idea also has tremendous implications for the classroom (and for any friendship). Play is one way for teachers to discover the "rules" of good teaching (and for people to discover/establish the "rules" of a relationship). It is a way for them to communicate the rules that will be used for learning to students in a non-threatening way (ebb and flow versus brick wall). It is also a way for observers to discover which teachers know those rules (at least subconsciously), which is not to say that one can always pinpoint which rules they know, just that they must have a general sense of the various factors that influence learning (which are complex because they cross so many different fields—neuroscience, psychology, subject area content, cognitive/social/emotional development, pedagogy, methodology, sociology, etc.).
The most satisfying forms of play require both the existence of an established relationship and a broad and deep understanding of the world—be that the world of a specific friendship, family, organization, community, or larger cultural group-though for some parties, the definition of "world" is both narrow in scope and limited in content). Additionally, play, in its richest forms, requires a substantial amount of trust because it demands a significant investment of self (often at the risk of great vulnerability—as is true anytime one chooses to deliberately break "the rules") and a willingness to purposefully put oneself into a state of disequilibrium—to be "off-balance" by consent.
That, ultimately, results in the following issues:
1) Is the climate such that it is "safe" to play?
2) Is there someone with whom to play? (An established relationship that spans at least one common world, founded on an agreed-upon and negotiated set of rules and common experiences?)
3) How "tamed" are the parties involved?
4) How likely is the "rubber band" of the relationship to snap back into its natural shape?
Those questions are not answered as easily as one might think—especially when it comes to grown ups. As St. Exupery's Little Prince once observed, "Grown-ups never understand anything by themselves, and it is tiresome for children to be always and forever explaining things to them. . . . Whenever I met one of them who seemed to me at all clear-sighted, I tried the experiment of showing him my Drawing Number One, which I have always kept. I would try to find out, so, if this was a person of true understanding. But, whoever it was, he, or she, would always say: 'That is a hat.' Then I would never talk to that person about boa constrictors, or primeval forests, or stars . . . . I would talk to him about bridge, and golf, and politics, and neckties. And the grown-up would be greatly pleased to have met such a sensible man."
So, I suppose that since I prefer to talk about primeval forests and stars, and the prevailing topics are golf and politics and neckties, I don't really have much to say as a general rule. Every now and again, though, I try my version of the Little Prince's "experiment." Sometimes it works, sometimes (often) it doesn't.