Tuesday, May 10, 2011

Four in the braid

Last week I went to a workshop at a thinktank in DC on social networks and behavior, desirable and otherwise. Something happened there. You will like it. Maybe not today, but soon.

Coming to the braid

As I participated in the conversations about social networks I did what I always do: I saw patterns. I noticed that the things people were saying about social networks could be described as having three dimensions:
  1. the flow of ideas, memes, concepts;
  2. who knows whom, relationships, nodes and edges; and 
  3. memberships, shared contexts, the life of groups.
So I started drawing little triangular and cuboidal pictures of these three dimensions interacting: ideas, relations, contexts. Then I noticed that they matched the "braid" of ideas I like to think about, which is based partly on Harrison White's three species of identity interaction. (If you know about this already, why not get a drink and come back. I'll just be a minute.)

In the second chapter of White's 1992 book Identity and Control (an excellent book for the patient reader) he explores three "species" of interaction between identities in human society, or ways people interact whether at the individual or group level (or both at once):
  1. selection, or sifting, sorting and categorizing;
  2. mobilization, or relating and persuading; and
  3. commitment, or working in groups toward common goals.
I have found these dimensions of great use in my understanding of how people interact ever since I read about them sometime around 2005. In 2009 I published a paper called "Three strands in a braid" about how they relate to aspects of identity and areas in the Cynefin framework and can be used to design better social software. There I put it this way:
  1. Selection of categorical aspects of identity based on characteristic–based evaluation of safety operates across the chaotic/complex boundary.
  2. Mobilization of relational aspects of identity based on membership–based evaluation of importance operates across the complex/knowable boundary.
  3. Commitment of positional aspects of identity based on placement–based evaluation of utility operates across the knowable/known boundary.
Or if you like pictures better:

So I was sitting there last week in this workshop drawing these diagrams and the braid came to mind. This was not surprising since we were talking about people interacting.

It's full of stories

But then, all of a sudden, another piece fit into the puzzle. I have no idea why it happened just then, because it should have happened six years ago, but I wrote this on the diagram:
  1. ideas = selection = story form
  2. relations = mobilization = story phenomenon
  3. context = commitment = story function
The explosion of possibility inherent in this connection may not be immediately apparent, but I have gotten more excited about it every hour since it first burst forth.

This table may help: (or not):

species of interactionselectionmobilizationcommitment
identity iscategoricalrelationalpositional
parts or wholes? each part singlypart to partpart to whole
evaluation is ofpurity, safety prestige, importance utility, quality
interaction space (these are White's terms, I don't like them)arena council interface
for examplemarketplace legislature aircraft crew
on-line examplematch.com slashdot.org SharePoint
the question asked isfriendly? hostile?dominant? submissive?useful? useless?
peoplemix and matchconnect and rankbuild and use
evaluation is bycharacteristicsmembershipspositions/roles
the activity issifting, including, excluding coalition building and dissolutionrole lock-in and ritual transfer
these things are networkedideaspeoplecontexts
the area on Cynefin ischaotic/complexcomplex/knowableknowable/known
hierarchy strength isuniformly weakrising and fallinguniformly strong
meshwork strength isrising and fallinguniformly strongrising and falling

Still here? Now let's add in the story dimensions.

species of interactionselectionmobilizationcommitment
the most relevant story dimension isformphenomenonfunction
stories havegenreshistoriesuses
example stories arefolk talesurban myths, rumorsmedical records
stories are studied bynarratologists, folkloristsanthropologists, sociologistscognitive scientists
stories work when theyattract, engage, stimulate, spark connect, travel, repeat, growexplain, reveal, surprise, overturn
stories help peopleexplore ideas, define themselvesshare experiences, build connectionsunderstand problems, make decisions
collected stories might be used to help peoplethink about what sort of person they would like to beget along with others, find common groundmake decisions and plans, solve problems
purposeful stories might be used toposition a productrecruit membersdesign a product

The reason this is so exciting is that it has implications for how we can deliberately work with stories to produce useful outcomes when people interact.

When I told my husband about this discovery he suggested a useful way to test it: try to break it. If I am right about my flash of insight, some combinations should make more sense than others. So here we go. To make the comparisons simpler I'll take the three examples of stories I give above (folk tales, urban myths, medical records) and think about how they might be used in selection, mobilization and commitment.

Testing story form: folk tales

Folk tales are used by all societies to make sense of life itself. They present a substrate on which we can explore categorical aspects of identity. Folk tales are often exaggerated, strange, obscene, frightening, funny and full of chance. They push us near but not quite into mystery. They don't tell us what to do or whom to trust, but they do give us a behavioral repertoire to mix and match. In any sufficiently complex body of folk tales one can "go shopping" for identity. I remember doing this as a child, wandering through books of fables looking for ones that "spoke to me" because they captured some essential element of my self-image. After I found a good match I would revisit it year after year to find validation in my chosen form. People do this today more often with movies and television and video games than with ancient folk tales, but the mechanism is the same: we create cloaks of identity out of characteristics assembled from the stories society offers us.

As people build their cloaks of folklore (also including songs and proverbs) they use them to represent themselves to others. My dad used to say, when he was working on some carpentry or mechanical project, "A blind man with diarrhea wouldn't stop to look at it." Meaning, it's good enough. But there is another message: I enjoy clever puns. (Shockingly, Google has not heard this joke.) This and many other silly little stories he used to tell were his way of expressing who he was. I use quite a few of his stories myself for the same purpose. Another person might say, "Good enough for government work," and that would say something about them (and they would know that).

Today many people use movies and television shows to think about who they are and to tell other people about it. If I say "that's shiny" I have identified an important characteristic of myself to a whole group of people. I even wrote it once on this blog, knowing it was a secret signal to other Firefly fans. I knew I shouldn't do that, I did! But I couldn't help myself. I wasn't reaching out to build a coalition with other Firefly fans: I was simply saying "Yes I am as bad-ass as the people on that show." Very silly and very human. In this sense folk tales and their contemporary counterparts are no different from the purchasing choices we use to declare our characteristics to others. We wear them, and because we wear them we try them on and see if they fit. Folk tale collections and movie schedules are wardrobes.

Are folk tales used to build coalitions? Not on their own. People sometimes use stories to convey messages and draw people to a cause, but most cause-related stories are not entirely fictional. I do have several folk tale collections on my shelves put together with themes - let me see - on girl power, peace, teaching, managing; but these are not used to draw people into a coalition. They are meant to be used by people already in the coalition, to explore how they will work within it and what it means to them. I doubt old men buy folk tale collections about girl power.

Are folk tales used to get things done? Sometimes, but obliquely. Story facilitators might use folk tales to help people build stories, but the folk tales provide only scaffolding, not content. They are subordinated to the real stories which are the more crucial element.

Testing story phenomenon: urban myths

Jan Harold Brunvald is the authority on urban myths. Says he in The Vanishing Hitchhiker:
In the world of modern urban legends [unlike in folk tales] there is usually no geographical or generational gap between teller and event. The story is true; it really occurred, and recently, and always to someone else who is quite close to the narrator, or at least "a friend of a friend." Urban legends are told both in the course of casual conversations and in such special situations as campfires, slumber parties, and college dormitory bull sessions. The legends' physical settings are often close by, real, and sometimes even locally renowned for other such happenings. Though the characters in the stories are usually nameless, they are true-to-life examples of the kind of people the narrators and their audience know firsthand.
If you have ever heard a real live urban myth, not a story about one, you probably remember it. When my sister and I went to college together she used to tell amazing stories to which I of course listened with perfect gullibility. She was sure they were entirely true and that some friend of some friend had actually heard them from a friend of ... and so on. It was only when I was much older that I found them in urban legend books and realized what she had been passing on. My favorite was the exploding toilet story, which I believed to be true for at least a decade. I fear I am too well informed now to experience the fun of participating in this societal phenomenon, but I think that's not unusual. The fact that urban legends move more through younger people is probably indicative of their greater need to understand how they fit into society.

The words in Brunvald's explanation all have to do with connection: recent, close, friend, casual, local, renowned, "the kind of people," firsthand. This connects strongly to mobilization, which has to do with the shifting connections of group formation and dissolution. Says White:
The council species, familiar in social mobilizations, are disciplines centered on a process of balancing contending but ever-shifting coalitions. Preexisting strings of dependency set up the endless process of mobilizing and remobilizing, as in an extended kin group with corporate rights whose allocations are balanced and rebalanced in a mutual discipline. The dynamics of contention keep this discipline going up and down in a scale of mobilization.
What he means is that mobilization is about lots of people working out who to trust and hang out with, which has something to do with who can get you the things you want and who might take the things you have. This is pretty much the function of urban legends in society. Here are the five most popular not-true urban legends on snopes.com right now, and what they say about groups:
  1. Putting your burned hand in flour can cure a burn completely. Message: the doctors are not telling us everything. Us: the people. Them: the doctors. (Note: This is not true. Don't do it.)
  2. Criminals are soaking business cards in a prescription drug called "burundanga" to incapacitate their victims. Message: People are waiting to capture you, and prescription drugs are dangerous. Us: the public. Them: criminals and drug companies.
  3. A movie is coming out soon that will portray Jesus as gay. Message: Atheists are out to destroy society. Us: Christians. Them: Atheists.
  4. A (fictitious) young girl is missing. Message: criminals are trying to steal your children. Us: parents. Them: criminals.
  5. Barack Obama was not born in the United States. Message: He is not one of us. Us: Americans. Them: Obama.
I was going to cover the top ten, but the top five made my point so well I didn't have to go on. Folk tales might differ around the world, but they are essentially the same stories about universal life adapted to local conditions. Urban legends might cover similar themes around the world, but they are locally situated in scope and meaning.

Are urban myths used to sift and sort? Yes, but that is not their dominant characteristic. There are collections of urban myths, but they are used more to explore societal movements and changes than to shop for characteristics of identity. People may build collections of stories they tell, but they are used more to warn people about other groups of people than to explore how we should live. When you get an email that says "Watch out because UPS uniforms have been stolen by terrorists" (the number 11 false tale in the current list on snopes.com) it is not about life lessons but about who to trust. Whom to trust, I know, but it just looks wrong.

Are urban myths used to get things done? The only scenario I can come up with that involves this combination is that of researchers paying attention to urban myths and rumors to find out what people are anxious about. However, that is not a case of the stories themselves being useful, but of the stories of the stories being useful. In other words, researchers pay more attention to transmission and change in the telling of urban legends than to the form of the stories (characters, setting, conflicts) or their function (cognitive benefits).

Testing story function: medical records

A strictly utilitarian set of stories, evaluated on quality for purpose, constrained to limited distribution, and rigidly structured, is found in the medical histories in the file cabinets of any doctor's office. (I mean, on the computer in any doctor's office.) My doctor's office has one on me: when and where I was born, my family history, my injuries and illnesses as a child, teenager and adult; my allergies, concerns, and lifestyle.

The primary use of medical records is to get things done: to treat patients effectively, consistently and carefully. Information usually does not travel between patients or doctors but stays close to its task. In fact, constraints on the spread of medical histories are crucial to the performance of the care team. If the patient cannot speak in confidence and have their history protected, treatment will be compromised. The cognitive functions of stories come to the forefront in medical records. Doctors look for patterns that help them diagnose conditions. Anomalies and surprises matter, as do correlations and correspondences.

Are medical records used to build coalitions? Only anonymously and never in detail. Research studies sometimes compile facts about patient demographics, lifestyles, and conditions; but they work with facts drawn from stories rather than from whole stories. Sometimes a group advocating for stronger funding for medical research on a condition will use the story of a patient to persuade people to join its cause. But the story used in this way is not a medical record; it is a personal, emotional appeal devoid of clinical detail.

Are medical records used to sift and sort? Not whole stories. Doctors and hospitals do sometimes compile information about patients in order to boost their diagnostic abilities. But as with the mobilization case, this information is generally compiled piecemeal. Even when people self-diagnose using stories told on the internet, they are not reading medical records; they are reading stories of personal experience, which include emotion and perspective but not detail. (Personal stories are not good choices to illustrate any of the extremes of story form, function or phenomenon, since they blend all three elements.)

Story dimensions and confluence

Assuming you don't violently disagree with these connections, let me explore a bit more into the confluence space of each of the story dimensions. To repeat:

species of interactionselectionmobilizationcommitment
evaluation is ofpurity, safety prestige, importance utility, quality
story dimensionformphenomenonfunction
example storiesfolk talesurban legendslife stories
hierarchy strength uniformly weakrising and fallinguniformly strong
meshwork strength rising and fallinguniformly strongrising and falling

Story form and confluence

If story form is most relevant in selection interactions, it should involve weak hierarchy and variable meshwork. In a collection of folk tales, whether it is written in a book or remembered by a grandfather, there is little if any hierarchy involved. Most of the books of folk tales on my shelves have no explanation for the order of the stories in them. A few categorize the stories by theme ("Animal Tales" and so on) but this is for convenience. What is important in folk tale collections is their flexibility to adapt to many connections and uses. An excellent storyteller is able to draw from their collection whatever story is most appropriate to any situation.

Divination systems, by the way, are essentially story collections, and they have little in the way of central ordering. The I Ching goes so far as to connect itself to the great meshwork of all being, which for about ten minutes in 1989 I may have actually believed. All divination systems connect themselves to some sort of order, but it's not hierarchy they call upon. The sad thing about divination systems is that because people think they ought to work, their great utility as books of wisdom is little known.

If story form is most relevant in selection interactions, it should place emphasis on purity and safety. This holds, because it is in the spheres in which story form is most important - narratology, screen writing, the study of folk tales - that the purity of form matters most. Books in these areas spend a lot of time on definition, categorization and exclusion. What sort plot arc does the story have? What motif? What genre? Is it a "good" story?

Here's a quote from a recommendation of a comic novel: "There's a purity of storytelling and a level of artistic proficiency here not seen often enough in American comics." Another, by an actor about a movie director: "'He's making a movie from his heart. Even though Chris makes these gargantuan movies, he manages to maintain that purity of storytelling. I think that's a big part of why so many people love his movies." Can you imagine these statements being made in reference to a rumor or a medical record? Or even to someone's personal story about their child's birth? Wouldn't that be an insult rather than a compliment?

Story phenomenon and confluence

If story phenomenon is most relevant in mobilization interactions, it should involve strong meshwork and variable hierarchy. This is true of urban legends, which are all about the hierarchies that control our daily lives. Government, doctors, pharmaceutical companies, food providers, police, organized crime, religious organizations, entertainment providers. It is hard to find an urban legend that is not about the actions of controlling hierarchies. You might think the legends about criminals manipulate meshwork connections, but no: they are about control, centralized control, by criminals over many people. The great majority of abductions are by family members, but these do not turn into urban legends. People who seize control of larger situations - serial killers, snipers, neighborhood burglars - are more likely to be found prowling in urban legends. Urban legends are all about control by people other than yourself.

If story phenomenon is most relevant in mobilization interactions, it should place emphasis on prestige and importance. This is also an element prevalent in urban myths, which swirl around the famous and infamous. When the people in urban myths are not well known or entirely made up, they are identified as noteworthy by their extreme appearance or behavior, or they connect to archetypally important images. Here's one: after the recent tornadoes in Florida, a story went around about a toddler found safe in a refrigerator who said, "A man with wings put me there." The prestige in this story is related to the power of angels. It is probably impossible to find an urban legend that does not relate to power as well as to group distinctions. The large number of current legends in the US about terrorists and Muslims shows how much people feel the need to work out who they can trust.

Story function and confluence

If story function is most relevant in commitment interactions, it should involve strong hierarchy and variable meshwork. This is true of medical records, which are centrally constrained in distribution and internal structure. You may manage to squeeze a personal story into your medical appointment, but your doctor is not going to write it into your medical record unless you mention an incident of a medical nature. Of course doctors think about similarities among patients and learn from experience, but the doctor is the central authority and stories cannot travel through any other path. Certainly patients do not share their detailed medical histories with each other; they rely on authorities to manage any coordinations. Variable meshwork comes in when doctors look over your file for connections between constituent elements of your history. Do you have a family history of heart disease? Do you live a sedentary lifestyle? Experienced doctors can see clusters forming in the attributes of a patient's history.

If story function is most relevant in commitment interactions, it should place emphasis on utility and quality. I've heard it said among doctors that a diagnosis can only be as good as the history that informed it. Medical schools place emphasis on learning how to take a complete and detailed history with high utility for diagnostic success. But doctors do not learn how to find out the social status of patients, which would be useful if mobilization was the goal, or whether they are dog or cat people, which would help to categorize and select them.

The quality of a patient history is so important to medical care that doctors sometimes direct their patients (or attempt to direct them) to keep symptom and food diaries. When I was a child I had terrible allergies and was hospitalized for intestinal bleeding that turned out be lactose intolerance. I well remember the arduous task of writing down every morsel of food I ate for months at a time (which of course made it impossible to sneak sweets). That's hierarchical control for you, and it chafes, doesn't it? Going to the hospital, may I never do so again, is not so much about pain but about loss of control. Memories of food control are probably why I refused to keep a migraine diary for decades, though I must of course sheepishly admit that finally giving in and doing this was what led me to the great reduction of the condition. The meshwork of connections between the foods and the headaches practically jumped out of the notebook and surprised me. The story had a function and carried it out. It had great utility.


So there you go. I think I have proven to myself that the scribbles I made last week do mean something. But have I proven it to you? Is anyone still reading this? No matter, I will plunge on.

Why does this correspondence matter? Because it has great utility on a number of different scales.

We can think about why different dimensions of stories work best in different interactions. We can tell stories with attention to those dimensions, and we can listen to stories with attention to those dimensions. Say I want to better understand how an organization works out issues of power and prestige. I should make sure I pay close attention to eliciting stories strong in the phenomenon dimension: perhaps rumors, or stories about rumors. And I should ask questions about those stories that pertain to story phenomenon. Where did this story come from? Has it changed? Where would you tell it? And so on. Now suppose I am planning a different project, one to help a product designer understand the fundamental characteristics of her users. Can you guess? I should pay attention to story form, because that will bring out characteristics for selection. I should probably get people to tell stories and derive their own archetypal constructs, as means of representing those characteristics in meaningful ways. Do you see?

We should also be able to think about transitions between dimensions in different parts of a story project. I did a project once where we asked inventors about the patent process. We had three goals.
  1. We wanted to understand what people needed to help them get through the process, so internal policies might be updated or corrected. For this we needed information on the quality of the process and where it might be improved. This was like medical records but instead of symptoms and spasms we wanted to know about bottlenecks and enablers. 
  2. We wanted to understand how people cooperated and connected as they invented things and worked on patent applications together. For this we needed information on the importance of various people in the system; who could get things moving, who people went to for advice. This was like finding out the urban legends about "how things work around here."
  3. We wanted to build a learning resource to help new inventors get up to speed. For this we needed information about the characteristics of an effective inventor, one more pure in their application of the process. We needed to find the heroes of the folk tales about invention and patenting and compile collections of wisdom.
This was an example of a project in which the goals were mixed; but the implementation of the goals took place at different times. At one point in the project we were writing recommendations to the people who ran the patenting process, so there we should have paid special attention to story function. At another point we were mapping out who we should invite to our story collection workshops, so there we should have paid special attention to story phenomenon. Finally, near the end of the project we built a learning resource that incorporated real stories. There we should have thought about story form. I hadn't thought of any of this back then and can't remember if I did pay attention to different aspects of story, but if I could go back in time I would use this idea there. Though I would probably go back to some other time, to be truthful.

Finally, this connects to something I've been thinking about for the past few months in relation to the field of organizational and community narrative. If the entire field of story work is a system, how could this "braid of four" help people work together to cooperate and build better projects together? My work lies in listening to stories, and this means I pay a lot of attention to story function and story phenomenon, but little to story form. Most of the people "on the telling side" pay a lot of attention to story form. I've always wondered how we could work together better, and this connection gives a possible answer. White and others have thought about what happens when interactions transition between the three dimensions, say when people move from selecting to mobilizing. We can apply these same transitional concepts to cooperation across different areas of story work. What should we pay attention to? What should we watch out for? I haven't got there yet, but I think this may be a promising path.

I am aware that I have probably not explained these connections very well. I have not yet figured out how to talk about White's dimensions in a way that doesn't cause people to look for the coffee machine. I half-wrote another blog post about where White's dimensions came from, because I think knowing their story makes them make more sense. I think putting it in this post would break something, so I'll post it later.

By the way, I was upset at first because my braid had grown another strand, but just in time I started reading the excellent book World Textiles, which explains that a braid can have any number of strands, not just three. I did not know that.

No comments: