Three-Toed Sloth

Books to Read While the Algae Grow in Your Fur, March 2013

Attention conservation notice: I have no taste.

John Levi Martin, Social Structures

The best approach to a theory of social networks I have ever seen from the hands of a sociologist. Specifically, it is about relating the content of different social relationships to the form of larger network to which they give rise, and how both form and content are linked to the ways participants think about the relationship — but not just to how they think. Martin draws very deeply on a huge range of scholarship, everything from studies of American teenagers at summer camps through the ethology of dominance hierarchies (including the origins of the concept "pecking order") to the history of European militaries and the development of party politics in colonial New York and Virginia. Astonishingly, he really pulls it all together, and writes much more than decently.

To over-simplify a lot, Martin wants to identify what allows some types of relationships to ramify into very large social structures, where the participants can nonetheless have some grasp of the structure and how it is organized. (It's not enough if the organization only becomes evident to an outsider after detailed sociometric analysis.) Principles like homophily don't work, because they would lead merely to cliques, or at best to dense clusters, and people are simply unable to handle dense social networks at any large scale; real networks are, and must be, sparse. "Balance" — the transitive closure of the idea that "the enemy of my enemy is my friend, and the friend of my enemy is my enemy" fails because it's strategic suicide. Relationships of exchange (of gifts, children in marriage, etc.) are more promising, but are also necessarily fragile, and hard to expand. Human beings simply don't do pecking orders, unless they are confined without possibilities of escape (like schoolchildren), and even then sorting a truly large pool of people by peck-ability doesn't scale.

The most robust possibility for creating large-scale, comprehensible social structures is an anti-symmetric patron-client relationship, grounded in some pre-existing inequality of status or resources (or ideally both), where clients offer services to patrons and patrons protect clients from other members of the patron class. If patrons can be clients of more important patrons, such ties can be concatenated into vast pyramids. In doing so, they do not lose comprehensibility (everyone remains a client of a single patron, who is superior on some recognized dimension), or call for dense networks, or for a carefully balanced flow of resources, or for vast mental efforts on the part of participants. Patronage is not a transitive relationship (my patron's patron is not my patron, whom I must serve), but patronage pyramids can as it were harden into command hierarchies, where subordination is transitive; Martin explores the role of this process in creating the modern corporation, army, political party, and state. (On the corporation, he is particularly good [pp. 236--241] on the fact that many people like being in control, and technical or economic efficiency be damned.)

To some extent, I feel that he leaves off just where things are getting interesting, by mentioning that in the case of parties, people create transitive relationships with each other by imagining that they have a binary relationship with the party, or perhaps with its ideology. Parties and ideologies, in this sense, are, to use the poet's words, "consensual hallucinations", but nonetheless very important to how any really large social organization happens. I wish this book said more about how they worked, but Martin seems to treat them as black boxes. (The phrase "imagined community" does not, to the best of my recollection, ever appear in the text.) I hope that this will be dealt with in a later book.

(Read on Kieran Healy's recommendation)

Update: "JLM" has a page of replies to reviews; the tone of the book itself is rather more staid.

Elizabeth Bear, Shattered Pillars

Sequel to the magnificent Range of Ghosts, continuing the action where that book left off. And there is a lot of action: escapes, betrayals, ambushes, storms, fires, eruptions, un-natural plagues, miraculous births; also cities (variously thriving, burning, seething with discontent, rebuilding and ruined), wondrous beasts, ghouls, forbidden books, slave poetesses, the intersection of chemical thermodynamics with wizardry, and cramped tunnels and vast skies. As an act of story-telling, what strikes me most, having just put the book down, is how much of the novel is told from perspectives of those who are (not to put too fine a point on it) villains, yet in such a way that the reader is invited to comprehend and even sympathize, though not approve. (Imagine key parts of The Two Towers being told from the perspective of Grima Wormtongue, who thought he was shouldering a burden by doing unpleasant things for the good of Rohan.) At the same time, Temur and Samarkar only grow on me as heroes. There will be a third book, for which I can hardly wait.

Elizabeth Bear, Bone and Jewel Creatures

Mind candy about dueling necromancers, but candy of a high grade. (How often, in mind candy or elsewhere, is the protagonist an old woman with arthritis?) In the same world as Range of Ghosts and Shattered Pillars, but thematically distinct.

Books to Read While the Algae Grow in Your Fur; Scientifiction and Fantastica; Networks; Commit a Social Science