Take a very simple Wittgensteinian language-game, e.g., a bricklayer says "BRICK" and the bricklayer's assistant brings her the brick.* All of this is part of the language-game: not just the utterance "BRICK," but also the assistant bringing the brick — so the actions as well as the sound. You don't have one part being language and another part not. It's all language, and if you leave out the actions it's not complete.**
Of course, at times the assistant could understand that he's to bring a brick, yet he chooses not to, in defiance or as a joke; or he may be prevented from doing so, say by an injury; and that doesn't mean the language-game is incomplete in these instances. As long as the practice is there, the established practice of "BRICK" and an assistant bringing the brick, the language-game is in effect. And defiance and humor are expressible in this language, too, even though the language only contains one word, the command "BRICK." (Suppose, somehow, there's miscommunication in the game. Or some misunderstanding, the assistant incorrectly thinking that it's only when the bricklayer has her arm raised as she's uttering "BRICK" that he's to bring the brick. Or maybe sometimes the bricklayer doesn't mean it, and the assistant has to figure out when. A game doesn't have to be conducted with absolutely certainty to be a game; a language doesn't have to have absolute certainty and consistency to be a language.)
We can define "language-games" as being, more or less, "human social practices." The terms "language-game" and "social practice" are near synonyms, language being so ubiquitous. But let's see what happens if we go further. Let's get rid of "more or less." Let's say that all human social practices are language-games, whether or not any word is actually spoken in the practice, and whether or not all the parties even know a language. Yes, at least one of them — the parent of a baby, for instance — will have to know a language; but the other(s) won't have to. So parental action and baby wails and goos and parental response are all in the category "language-game." A baby being initiated into parent-child social behavior is a baby being initiated into language.***
By this definition, all musical events, including the "nonverbal," are nonetheless in some language-game or other. This doesn't mean "can be made part of a language-game by translating musical sounds into words or by describing the music in words." It means that the language-game includes musical sounds as they are, and we can take the sounds and see their role in particular games — particular social practices — just as we can take the utterances and actions in the "BRICK" language and see their roles in that particular practice. In any event, we refuse to give the social practices we call "music" the special status of being "nonverbal." They aren't.
( Motive here is to tease out what might be usable in Mark's glimmer of an idea )
( Footnotes (as opposed to musical notes?) )
Of course, at times the assistant could understand that he's to bring a brick, yet he chooses not to, in defiance or as a joke; or he may be prevented from doing so, say by an injury; and that doesn't mean the language-game is incomplete in these instances. As long as the practice is there, the established practice of "BRICK" and an assistant bringing the brick, the language-game is in effect. And defiance and humor are expressible in this language, too, even though the language only contains one word, the command "BRICK." (Suppose, somehow, there's miscommunication in the game. Or some misunderstanding, the assistant incorrectly thinking that it's only when the bricklayer has her arm raised as she's uttering "BRICK" that he's to bring the brick. Or maybe sometimes the bricklayer doesn't mean it, and the assistant has to figure out when. A game doesn't have to be conducted with absolutely certainty to be a game; a language doesn't have to have absolute certainty and consistency to be a language.)
We can define "language-games" as being, more or less, "human social practices." The terms "language-game" and "social practice" are near synonyms, language being so ubiquitous. But let's see what happens if we go further. Let's get rid of "more or less." Let's say that all human social practices are language-games, whether or not any word is actually spoken in the practice, and whether or not all the parties even know a language. Yes, at least one of them — the parent of a baby, for instance — will have to know a language; but the other(s) won't have to. So parental action and baby wails and goos and parental response are all in the category "language-game." A baby being initiated into parent-child social behavior is a baby being initiated into language.***
By this definition, all musical events, including the "nonverbal," are nonetheless in some language-game or other. This doesn't mean "can be made part of a language-game by translating musical sounds into words or by describing the music in words." It means that the language-game includes musical sounds as they are, and we can take the sounds and see their role in particular games — particular social practices — just as we can take the utterances and actions in the "BRICK" language and see their roles in that particular practice. In any event, we refuse to give the social practices we call "music" the special status of being "nonverbal." They aren't.