A recent post to asfaculty suggested that interpersonal "spats" be kept private and this started me thinking about information flow in organisms (of which an "organism" - club, government, university is a single sub-category).
As a computer person I have a tendency to try to see things in the kinds of formalized, abstract models that might be used in software constructions that mimic structures in real life. Some time back my interest in artificial life and self organism (especially in comptuting structures) started me thinking about organisms and how they work - whatever an organism is. And I've come to believe that one interesting model of organisms reduces them to flows. These flows are of two main types, one is physical (people, and things) and the other is information. It's not always easy to separate these two flows - in the brain, for example, the physical flow consists of chemicals and electricity and that physical flow serves as a conduit for the information flow. Money is another good example of the difficulty of this separation - while money does have a physical form (bills, coins) it is also abstract and may be represented only by bits in computers. Money is still more "physical" than many other kinds of information which may be so ephemeral as almost to be unmeasurable (one on one voice communication is an excellent example). Technologically, there are ways to send information that are designed to be almost indectable. (Steganography is one of these.)
In studying human organisms (ranging from societies to clubs to universities) it might be worthwhile to determine the kinds of "information flows" involved and measure them - certainly one of these is money (budgets), one might be people movements in the hierarchy (that is, not necessarily physical), and one would have to be person(s) to person(s) communication.
I've wanted (mostly out of curiosity) for rather a while now to get some way to measure all the communication flows in an organism (like a university) and to diagram it all in a map of sorts. To to this well would involve trying to measure every email, every memo, and every conversation (in person or on the phone). Not very practical, sadly.
It seems quite possible that we would discover, given enough such data, that the organism does not work the way we think it does. I conjecture that there are "information centers" through which almost all important information flows and which serve as the real control mechanism of the organism. Such information centers might be individuals or small clusters of people with very high information interchange internally and widespread information connections externally.< /p>
Without having any formal role in the organism, such information centers can easily have large effects on the organism. One example I've seen was a secretary for a higher level manager with a large formal role. This secretary had informal, but close, contacts with many other people in the organism, and had as well, quite a bit of power over what the manager saw and when. Effectively this person could (and did) lower the priority of contacts between the manager and others, and could (and did, however indirectly) have a very large effect on the policies of the organism.
This is not necessarily a bad way to run an organism - in fact, it can be very effective, especially in small organisms where frequently there is a single nexus and almost everyone involved is at most one or two communications steps away from the nexus.
It might help the non-mathematically inclined to remember the notion of < a href="http://www.newscientist.com/ns/980606/nbacon.html">"six degrees of separation" (or Kevin Bacon) (and here, here, here) - the idea that any two people in the world are linked by some chain of six people, with any two consecutive people in the chain knowing each other personally.
Every basic communications channel is noisy (in a technical sense), and most real communications channels are composed of multiple basic ones - and worse yet, the links between the channels themselves introduce noise into the process. The further the communicating partners are apart (in terms of the length of the communications channels), the less reliable the information (the childrens game of "telephone" is a good example. In an organism, the further from the various communications centers something is, the more noisy and less accurate the information they get becomes. (There are ways to produce more accurate communications links, but in a careful information theoretic sense, there is no effective way to produce a perfect link.) For persons far enough away (in terms of communications links, not physical distance), the information available elsewhere on the network can become effectively inaccessible. (For example, during his lifetime, it was widely known in some circles that J. Edgar Hoover was a cross-dresser - but this was almost entirely unknown to most of the people in the US.) Communications channels, it should also be pointed out, are four dimensional, sending information not only over spatial separations, but also over temporal ones. And in this context memory itself is a communications channel, and the "absent minded professor" stereotype should drive home the point that memory is a very unreliable way to transmit information into the future. (More interesting is the Sapir-Whorf hypothesis about the structure of language and the workings of the brain - it seems more than likely that the very act of framing information in linquistic terms imposes a systematic (and perhaps language dependent) lossy transformation of that information.)
Probably the most effective way to counter this degradation of information quality over communications channels is to communications links that keep the degradation of information to a minimum and make it available to outlying (spatially, socially and temporally) parts of the organism. Many human institutions do this by creating new information centers to distribute information more widely and reliably. There are many ways to do this - archives (libraries, journals of record), mailing lists, wide distribution memos, the world wide web, committees with reasonably disparate memberships (it does little good in terms of information distribution to build a committee that reflects an information nexus already in place).
Without even creating archives and other centralized information centers, a very good job could be done by just introducing new random communications links. For example, adding random people to a committee, building a mailing list that sends messages to random people in the organization. This process alone can reduce the "six degrees of separation" to some smaller number. There was a very good article in the New Yorker a while back on just this topic - might anyone have a copy I could steal (or copy) ? For those capable of writing small programs its interesting to build a graph with a high diameter, then add random links and watch how quickly the diameter falls. For those more mathematically inclined, this process reflects the growth of the giant connected component in random graphs.
It must be emphasised that these added links are still noisy, but since the links are now independent and the total length of the communications chain is shorter, the information does not degrade as quickly.
Why should anyone care? Degraded information can lead to instability and problems within an organism - but paradoxically, it seems that sometimes it can also lead to increased stability. When reliable, low noise communications channels are not available, one major risk is that an organism of any size will develop several of these communications centers which will usually interact only weakly. This is a pretty good way to get internal fragmentation and dissention. (Politically, it seems a reasonable hypothesis that most revolutions take place when exactly this happens.)
In the asfaculty posting, it was claimed that publicly posting information about interpersonal communications problems was A Bad Thing. This interested me quite a bit as it seems to suggest that only "official" information about "official" matters should be shared in public fora. Perhaps only information approved by committee is appropriate. Private discussions, though, are very important, even essential - they are a natural, very human mode of interaction, maybe even the best mode of interhuman interaction. They are also important to organismsn as they allow ideas to be considered informally and without committing the organism to any policies.
If the only discussions are informal and one on one, then persons farther away (in a communications sense) may never have the opportunity to contribute and may never even hear about things. Since these people may have valuable and relevant information, this is bad for the organism. at input. On the other hand, if all discussions are formal and wide-distribution, they become speechifying or worse. Watch CSPAN for a bit, or think of the kind of canned "PowerPoint" style presentations we've all endured, which usually provide the appearance of information with little or no true informational content. This often results in the true discussions being forced behind closed doors and enforcing a kind of informational have/have not dichotomy.
Suppose, for a moment, that there is a small coterie of individuals connected by high quality information channels who are using these connections to impose their personal values and visions on an organism. If alternative information channels are suppressed (deliberately, structurally or otherwise), such a group can, in the right circumstances gain quite a bit of unmonitored power. Indeed, we don't need to look far to see such things - Watergate, Stalinist Russia - and many more examples come easily to mind. (I don't tend to believe that such coteries are deliberate "conspiracies" as I agree with the person (unknown to me) who said "Anyone who believes in conspiracy theories is seriously overestimating the human race", and because I believe that it is easier to explain the same kinds of behavior with theories that require substantially less stringent assumptions (nods to that Occam guy again).) Several times since I've been here at Eastern I've seen cases where it was clear that communications were failing. The strategic plan and the general education process are probably the most widely known, but there have certainly been many others.