I'm working today on my chapter about how we are socialized into our roles in information networks. It's called "Learning to Be a Node." The section I'm working on is about the use of children as spies. Huh? What this refers to is how up to a certain age kids are informationally unsophisticated and are treated as such. They can go "back stage" (Goffman 1959) and no one even thinks about it. If they reach the age where they can observe and report on their observations before they learn what one does and does not talk about "out there," then outsiders can take advantage and use kids to peer inside the family unit.
The first example is the largely apocryphal case of Pavlik Trofimovich, a Russian boy who, in the 1930s, is reputed to have been killed by his family after he turned his father in for anti-communist behavior. The next examples are the case of the Romanian secret police apparently using children for similar purposes during that country's communist era and the attempts in the 80s and 90s by anti-drug zealots in the US to get kids to call 911 at the first sight of anything related to drugs in their own homes (that is, to turn their parents in to the police).
Those examples are all about how the state can use kids to pierce the information firewall that surrounds the family. The last example comes from a 1967 anthropology article in which John Hotchkiss describes how villagers in Chiapas, Mexico, employ their own kids as roving information gatherers to learn what's happening in other families. The kids are given free reign and adults act unguardedly around them so that their own parents can ask them, when they come back from errands or from visiting with friends, "what's going on over there? Is Sr. Gonzales still drinking?"
Of course, the kids are potential double agents. One of the reasons that adults are unguarded around them is because they can also pump the visiting child for information about her own home. Hotchkiss is primarily interested in the ways that the kids act as intermediaries that permit information to flow and for material transactions to occur without the adults losing face (cf. Goffman 1959). It serves my purposes, though, because it shows how, until we are socialized into the notification and information norms of each network, we are sources of potential "leaks" by spilling beans that we don't know are beans (and, correspondingly, the sources of blindspots by not passing along things that we ought to).
You might detect an undercurrent here that will connect this chapter to the one on notification in organizations: the whole question of how to build an effective intelligence network so as to avoid "stovepiping" and related problems that were said to be behind the intelligence failures that led up to 9/11.
Goffman, Erving. 1959. The Presentation of Self in Everyday Life. Garden City, New York: Doubleday Anchor Books.
Hotchkiss, John C. 1967. "Children and Conduct in a Ladino Community of Chiapas, Mexico." American Anthropologist New Series, Vol. 69:711-718.