If You Can’t Hear Anything Nice, Don’t Hear Anything: Robustness vs. Civility of Networks

John Kelly/WashPo, "Radical Civility" 2009

John Kelly/WashPo, “Radical Civility” 2009

A call for online civility, particularly on Twitter, was recently issued by Kathleen Fitzpatrick, Director of Scholarly Publishing at the Modern Language Association, in her blog post “If You Can’t Say Anything Nice“:

Folks, we need to have a conversation. About Twitter. And generosity. And public shaming…How can we develop modes of public critique that are rigorous and yet respectful?

I appreciate what Fitzpatrick is saying, and agree it would be good for people to generally have more generosity and consideration in communicating. Also, Fitzpatrick, whom I’ve had the pleasure to meet and discuss scholarly-communications issues with, is I’d say rather exemplary for generous, open, and forward-thinking discourse.

(also, disclosure, she has joined the Academic Steering and Advocacy Committee for the Open Library of Humanities, of which I am a co-founder. This actually happened while I was writing this post, that was pretty cool. Thanks Kathleen! Check’s in the mail).


“All Our Grievances Are Connected.” Occupy NYC poster

On the other hand — to venture a bit of public critique that I hope is respectful — I believe there are significant dangers in asking others to behave as one would like, or as one believes oneself to be behaving, or trying to draw a line between, say, as Fitzpatrick does, “thoughtful public critique and thoughtless public shaming.”

The problem with asking us to behave the right way towards each other — the golden rule of reciprocity — is that, to paraphrase Kant, from the crooked timber of humanity, no symmetric thing was ever made.

Kasson, Rudeness and Civility, 1990.

Kasson, Rudeness and Civility, 1990.

To start, who are we calling “we”? There’s a history of dominant-group practices being asserted as natural “manners,” “civility” or such concepts. While people commonly say things like “basic civility” or “well-mannered,” historians and anthropologists such as Edward T. Hall tell us that interaction behaviors vary greatly across and within times and cultures, and can change fairly rapidly — for example, the mid-19th C. erosion of obligation to speak when addressed by a stranger on a city street, as examined by Richard Sennett in The Fall of Public Man. We can see today that attitudes, for example, on appropriate mobile-phone use, texting, or multitasking vary wildly around the world and between Americans.

LRB-coverIn my own work recently, looking at various online knowledge communities, it sometimes seems that different fields communicate so differently as to be almost mutually incomprehensible. For example, the savvy/snarky Silicon Valley engineering banter of Y Combinator’s Hacker News, vs. the genteel belles-lettristic London Review of Books, vs. the roving criticality of the The Disorder of Things. What is a deft critical sally or pertinent reminder in one place might easily be blunderbussing, precious, or preposterous in another.

Secondly, an insistence on manners can often be a way to evade the discomfiture of having one’s thinking or practices challenged. This point was highlighted in a recent talk by legal scholar and civic activist Lawrence Lessig at Stanford, in which he argued that the political reform critically needed in the U.S. actually requires us to challenge and “make uncomfortable” our often like-minded acquaintances, to break the local convergence of opinion that creates crippling national polarization.

Applying this line of thinking to the scholarly context discussed by Fitzpatrick, I would suggest that academic collegiality might be defined as the ability to have the widest possible disagreements — therefore, demanding from others the fewest rules and least concordance possible. To be other-engaged, and world-engaged, we need to be always leaning in to the uncomfortable. After all, the world is mostly other people who aren’t like you.

Paul Gavarni, Le Flâneur, 1842.

Paul Gavarni, Le Flâneur, 1842.

Consider this just from a practical standpoint for a moment. We live usually in cities, environments of complexity hardly conceivable 200 years ago; our lives are constantly woven together from a plenitude of networks and contexts. The more fleeting and dissociated and variegated our interactions — in the global Baudelairean “City of Modern Life” or the global glorious noise of Twitter — the more opportunities for civil symmetry to fail due to cultural difference, context collision, noise, or accident. Smooth functioning with diverse practices requires a large granting of slack, or its less-inspired cousin, mutual indifference. As Baudelaire dryly observed, “It is only by universal misunderstanding that all agree.”

Speaking of network behavior, this longstanding cultural observation above actually underpins a core Internet engineering idea, the Robustness Principle. This is also known as Postel’s law, after Internet pioneer Jon Postel, who wrote in an early specification of the Transmission Control Protocol (TCP) that:

TCP implementations should follow a general principle of robustness: be conservative in what you do, be liberal in what you accept from others.
(RFC 761, 1980)

So for example, you should send documents that you believe to be strictly compliant with standards, but accept documents that diverge from standard in common ways.

Jon Postel, formulator of the "Robustness Principle" in RFC 761

Jon Postel, formulator of the “Robustness Principle” in RFC 761

To put it another way, the Robustness Principle suggests that exerting common rules (civility?..) may actually *prevent* a “society” of agents from working well, because it’s prone to be fragile, unbending. The Internet engineering culture, while of course heavily about peer-to-peer civility, leans strongly to the forgiving and capacious notion of how to get along, and away from the regimented. (“rough consensus and running code”).

This freewheeling engineering culture might be viewed as a sort of Taoist, tree-that-bends-doesn’t-break antithesis to the Confucian, ordered-society Bell System network culture before it.

from "The Social Network" (2010)

from “The Social Network” (2010)

Incidentally, to psychologize irresponsibly for a moment, one suspects that many key figures in computing/network culture had some personal experience with illiberality: being at some odds and frustration with whatever was thought civil and sociable and natural, and perhaps not knowing how to speak/behave acceptably. 

Mark Zuckerberg trying to crack the Harvard social codes, for example. 

Among the reasons that common rules for civility may fail, aside from randomness and noise quotient, are the many ways we don’t even understand ourselves. Cognitive pyschology has shown how systematically people mis-perceive and mis-estimate their own behavior — for example, the near-universal phenomenon that most people consider themselves above-average in many traits, a logical impossibility.

Even if people do perceive their behavior accurately, they can’t necessarily change it — for example, many aspects of face-to-face communication are unconscious and sub-perceptual, difficult or impossible to change. Finally, even if you are perfectly self-aware and controlled, you can’t know how other people will perceive you.

So, I fear for the richness of our conversations, if we’re enjoined to not say anything unless we know we’re saying something nice. Better, perhaps, to observe Postel’s Robustness Principle, and keep speaking, speak well, but also listen nicely. If you can’t hear anything nice, just try to hear nothing at all — and this way the open network leads.


[thanks Katina Rogers for pointing out a grievous thrice-repeated typo, fixed now.    1/26 3;15pm pst].