The recombinant DNA of our political discourse

by Carl Dyke

I’m forging slowly and painfully through John H. Holland’s Complexity: A Very Short Introduction (Oxford UP, 2014). I’m going to recommend it, but it’s tough going for me because Holland comes at complex adaptive systems from an engineering and computer science perspective. The approach is computational, so I would imagine tremendously congenial to Asher and perhaps JohnM. Whereas I can work it out (thanks, John), but it’s way out of my silo. I’m a house painter, not a pigment chemist, Jim. Consider a passage like this, from Chapter 5, which took me about an hour to decode:

To get a more precise view of the routing possibilities for tags, let’s look again at the use of #s to define the conditions for signal-processing rules (introduced at the beginning of Chapter 4). A condition can be set to respond to a given tag by using #s (don’t cares) on either side of the tag; thus the condition #100### … # responds to a signal string with tag 100 at the 2nd, 3rd, and 4th positions of the string. Note that a condition with many #s can accept a variety of tags while a condition with few #s will accept few if any tags (requiring a specific string when it has no #s). Or, looking at tags themselves, short tags satisfy a variety of conditions, while long tags make highly specific requirements on conditions.

“Don’t cares?” Wah? And then he does some tables and stuff to show how the routing and recombination of tags happens, creating emergence. Which normally I’d ignore and wait for the explication in words, only then the words assume I have a basic understanding of how binary works; and although I do, I also kind of don’t. My balls are not making it through the urn gates, if you get my drift.

“If you’re not bringing math skills to the problem [of climate change],” [Bill Gates] said with a sort of amused asperity, “then representative democracy is a problem.”

So as usual I resort to a kind of rough and ready translation, to find enough cognitive overlap with something familiar where I can get a handle on what I’m being told. Which, if I understand correctly (but how could I, given this cludgy workaround), in itself corresponds to what Holland is saying about how information tagging works, and the emergence of something like knowledge. And because of the election and the work I’ve been doing to try to bridge the various discourse communities I interact with, the something familiar I go to is politics.

I’m especially interested in, roughly speaking, the information content of political discourse. I surf a lot of short-form information streams, ranging from domestic and foreign press and commentary (Atlantic, Guardian, Jacobin, Baffler, Breitbart, Red State, etc.) to online affinity groups (Sanford / Lee County Truth Seekers) to Facebook statuses. I interview Trump ralliers for my department’s oral history project. I talk with neighbors and friends. None of this is meant to be scientific or in any way dispositive; I’m just grazing.

What I find is that American political discourse seems like it could be usefully analogized to DNA. There isn’t actually that much divergence or diversity of original information. You can get almost anything said by rearranging and elaborating on a few basic concepts like individual, community, threat, and opportunity. DNA is just four proteins, CGAT. That’s plenty. How they’re arranged is what actually does the work, all the way up to thoughts about Life, the Universe, and Everything.

Analysis of complex systems almost always turns on finding recurring patterns in the system’s ever-changing configurations (Holland).

There’s nothing new about ‘reducing’ political discourse to unit ideas. Recall George Lakoff boiling American politics down to basic orienting metaphors of the family: authoritarian father and nurturing parent. But those are already speciated information complexes, and as the subsequent commentary has shown, highly vulnerable to complexifying analysis. He’s missing the rearrangement step. It’s like if he said, “Over there you have monkeys, whereas here are the humans.” And because the matter is in motion, just about everyone turns out to be a missing link.

It is unusual for CAS agents to converge, even momentarily, to a single ‘optimal’ strategy, or to an equilibrium. As the agents adapt to each other, new agents with new strategies usually emerge. Then each new agent offers opportunities for still further interactions, increasing the overall complexity (Holland).

 So here is a human interaction I wrote about on Facebook the other day:

I don’t suppose I’m hiding my disdain for Trump and (qualified) support for Clinton very well. And yes, I think there is a lot of bad stuff you have to ignore or explain away to be excited about either of them, but so much more so with Trump that there’s no reasonable comparison.

But the world keeps turning. Yesterday I was clearing two big trees that fell across our driveway when two pickups pulled in and spilled out neighbors we hadn’t met yet, equipped with big chainsaws and neighborly fellowship. A job that would have taken me an hour took us ten minutes, and then off they went looking for more people to help.

The conversation made it clear they were Trump supporters – pretty much all our neighbors are – but I liked them very much and not just because they were helping me. They were no nonsense and capable, thoughtful and droll.

It sort of amazes me that good folks like this can’t or won’t scale their sense of helping out to the national level. I know that who needs help, and when, and by whom, and for how long, is a complicated question. Why people need help is even more complicated. Unless it’s a tree across the driveway. On that we can agree.

What does this agreement mean? Holland says that what I’m going to call understanding, which is a transformed relationship to the world, comes from the kinds of overlap that enable exchange, and that change emerges routinely from these swaps.

It has been observed that innovation in CAS is mostly a matter of combining well-known components in new ways (Holland).

So one overlap in my story is “chainsaw,” and it’s actually a critical one for this particular interaction. There’s even political information of a kind in it, since owning your own chainsaw and being usefully skilled with it express a position in and relationship to the world. These guys got a lot more comfortable once they saw me whacking off limbs and logs without a lot of fuss, and vice versa. This overlap then enabled later bantering swaps in which we explored (scant) overlaps in our gender and family systems via their assumption that my hatchback, trapped behind the trunks at the bottom of the driveway, must be my wife’s car. The rescuing the little lady narrative got weird for a second. It got ok again once it was the commute vehicle, and a diesel, and we could talk about retrofitting an electrical fuel pump on old Jettas to improve reliability.

The guys were grandfather, father, and son. Grandpa mostly sat and watched, Dad ran the main saw (he said they had “bar and bow,” which I didn’t properly understand until he pulled out his rig, learning and growing),


and Son sort of puttered around moving cut branches. In the following conversation, Dad talked modestly about a life of extraordinary accomplishment as a builder, mechanic, and entrepreneur. Son had the same easy feel of quality but turned out to be one of the completely useless young men I’ve written about before. He’s got a degree in History, so we overlapped there, and on our assessment that no one in their right minds would go into teaching in the current educational environment. Just a whole lot of overlap there, to be honest. He noticed my Farm Bureau Insurance cap and connected it to his pragmatic pursuit of insurance agent certification. There were no illusions about the redeeming value of this occupation, and here Dad made a crack about getting right with Obamacare and the mirage of “free” insurance, which is one of the branching tags of our current politics.

OK, so this is a blog post and can’t noodle around in anecdote forever. What about the recombinant DNA of our political discourse? The overlapping tags for us were the American classics: community, individual, opportunity, threat. In some sense we agreed on all the basics; that is, our ideas were composed of the same elements. Our political divergence was in those tags’ recombination across different environmental and developmental histories. In the right situation we could get the community and opportunity tags to overlap pretty robustly, but what became clear is that for a whole range of situations, that was not the effective exchange zone, branched especially by where we put the threat tags.

Roughly speaking and allowing for all the partial overlaps, conflations, and plain misses I think this is basically right, but I’m no more sure it’s usefully right than noticing that monkeys, apes, and humans share 98% of their DNA, or whatever it is. As Holland might say the vocabulary needs the grammar to emerge into meaning. On the other hand there’s always that idea of kinship that’s so easy to lose in our biological and political discussions. It’s not that conservatives don’t have the individual tag, it’s that it’s assembled differently with the community and threat tags. We could work with that, maybe. And as usual this is me thinking out loud where, if we can get the tags to overlap, others might be interested to think with me.


6 Comments to “The recombinant DNA of our political discourse”

  1. I like turtles.

  2. Merda. This piece blew the cork out of of the jug where I had a lot of things stashed that I didn’t want to be thinking about for a while, and I can’t desist. Fortunately, I’ll forget half the stuff while I’m doing the other half, so this won’t be near as long as it could be.
    First, of course, the disclaimer. We’re in pretty much the same boat with respect to the formal aspects of all this. I probably know more of the math, you definitely know more about the computational stuff. In particular, you’re fluent with it at a practical level where I;’m hopeless (I can’t even find my DV dashboard). In contrast, I can handle Poincare kinds of stuff. This makes us, notice, yet another case in point for the point you want to make. Fortunately, we both know Lakoff pretty well. We also know the appropriate episode of Next Generation. So I can say, let’s compare #s with “with sails unfurled”: tags with imagery.
    Holland stands alongside Deacon in the development of our understanding of emergence. He’s also a long step closer than Deacon to the digital canon. For Deacon, DNA is a constituent of biochemical processes; for Holland DNA is a stock of digital atoms involved in processes of combination and recombination. Deacon wants to know how you can get new meaning through causal process; Holland wants to show that you can get new meaning through recombination. In both cases there’s a necessity to keep track of the programs that are running in the background — programs that they can make invisible for their primary constituencies, and opaque to us. Holland’s program is to keep as close to logical atomism as possible, while overcoming the limits of pure additivity; Deacon’s program is to re-evolve the biology from the molecular level. Without arguing it at length, I’ll say that in both cases meaning has to arrive from confounding a top down with the bottom up, and hoping that the top down can be shown to be (naturally) present to “manage” emergence. In both cases. To put it in other words, structural matrix has to be there from the beginning.
    So, first, I’ll buy the social DNA analogy only if it isn’t stripped to the level of the bare particular. That is, I insist that the epigenetics, regulation machinery, interaction with environment, and so on be part of the analogy.
    Second, I’ll hypothesize that a major part of the structural matrix in the social case is imagery — or whatever Lakoff wants to call it. From there we can move on with sails unfurled.
    Third, I can buy the perfect aptness of your chainsaw narrative. As you can imagine, I have close cousins to it. In particular, for present purposes, my search for people to buy things from, people to fix what I need fixed, and people to chat with in the back field results (accidentally?) in collecting supporters of Donald Trump. We share the image of candidness and “rugged individualism” lamentably threatened. The move from #s to the sails unfurled marks a bifurcation in structural and functional complexity of symbolic space. It too has emerged, but in its emergence it hasn’t left less complex emergence behind.
    Now let’s move to the complex juxtaposition of cooperation, persuasion, and live-and-let-live in your narrative. I’ll deal only with the distinction between changing minds and supplementing minds — subtraction processes and addition processes — thereby plunking us down in our classrooms. You can easily clarify that cryptic fog for yourself. I’ll just add a narrative to yours.
    As you well know, I’ve spent the summer gaining entry into the local art world, and (modestamente detto) been spectacularly successful in doing so. So last Wednesday I spent two hours at the Phillips Mill Art show as the artist in residence. (All the people who have had a piece accepted are gently pressured to do so.) The Phillips Mill Show is the prestige topper of the local art season. Among other things, the gig allowed me a chance to explore the process of selling my work: face to face with a potential patron. So I was there as the sculptor of “Nuts”, and a natural object of curiosity. So I was sought out and probed. In particular, a group of four or five women came along, said some very complimentary things about Nuts, and asked some intelligent questions about it, woodcarving and so on.
    The local art scene is dominated by a few painting traditions strongly tied to local ambiance and history. Pennsylvania Impressionism is a known and honored genre, and, for example, there are big prizes especially for works in and around that tradition. But the ladies loved my piece. It’s unfamiliarity was actually an asset. Its beauty was fully appreciated. My job was to translate that interest into the world of credit cards and checkbooks.
    Well, I’d brought along a couple of show-and-tells: one of those little spheres with a ball rolling around inside, and a ball and chain, both in cedar. So I handed them around. One woman in particular couldn’t take her hands off them. (I refrain from talk of caressing my balls.) The smoothness of texture was an aesthetic virtue. In fact, the whole conversation broadened the women’s sense of aesthetic quality. They were good at showing this. Good analogies were made, etc. At his level of the game I think I added to the scope of their aesthetic sensibilities, without subtracting anything from their previous sense.
    But there are the checkbooks. Am I actually serious about selling the piece? If so, then, supposing that they were really potential buyers in the first place, I have to shift their budget priorities away from barns and bridges. I have to convince them that a moderately messed with log is an artwork in the same way that a bronze casting is (but a lot cheaper), and a live candidate to supplant an Autumn day in Carversville at the top of their budget. So replacement joins supplementation. I don’t know yet if I did. But, sale or no sale, did I install a new imagery for “worthy work of art” to go along with the previously entrenched ones?
    The artworld is a really convenient microcosmic model — and, at the dollar top of Sotherby’s et al more than a microcosm — for examining the dynamics of choice. Because i wasn’t asking anybody to renounce their faith in Pennsylvania Impressionism, or anything else, mine is a gentle case. Real political choice, these days, is not often a realm of gentle cases. Here lies the power of your chainsaw narrative, the narrative of preserving the space of gentle choice. When subtractions swamp additions, the choices start to get nasty. All the difficulties of pluralism lie there. All the difficulties with climate change lie there.
    Ironically, most of the processes involved in the development from zygote to adult can be looked at as delicately orchestrated subtractions. But that’s for another day.

  3. I grabbed a copy of Holland. I’m interested to see if the explanations come easier to me. I assume they will, since it sounds an awful lot like Holland’s old classifier work of yore.

    A couple of very quick, undeveloped thoughts:

    In the biological metaphorical space, genes can deactivate other genes. I think the idea is present in Lakoff – it definitely is in Haidt – that we generally have the same tags, but they’re not just arranged — they’re also prioritized. And the prioritization doesn’t always seem to fit with a metaphor of “ranking by strength” wherein the tags at the top are just stronger. It often seems that high priority tags simply deactivate lower-priority ones.

    And speaking of deactivation… the virtue of the driveway is not easily extended to the larger world maybe because of a line that it seems we all have in our heads wrt seeing things as human or not. This line is a lot more powerful and a lot more easily tripped across than I think it’s given credit for, and there’s a *lot* of rhetoric out there that, consciously or no, is trying to shove us across it. Like our priorities, the line acts more like a switch than a graded force.

  4. Try Jean G. Boulton, Peter M. Allen, and Cliff Bowman Embracing Complexity: Strategic Perspectives for an Age of Turbulencefor a more accessible introduction. Then, if you are inspired, try Melanie Mitchell Complexity: A Guided Tour.

  5. Thanks for the good reading suggestions John – everyone note. I also recently enjoyed Donella Meadows’ Thinking in Systems, and to me the essential classic remains The Evolutionary Dynamics of Complex Systems.. In case you’re worrying about me, of course I’ve already read a whole bunch of the easy stuff, that’s why I’m moving on to the hard stuff. I don’t trust that what’s easy to me is therefore good, and I also want to be in on more conversations, even if I’m the guy in the clown nose.

    Asher (and DtE) yes, Holland keeps it pretty close to the basic operations and so he’s vulnerable to dismissal as a reductionist. It’s not at all obvious how or whether to scale his analysis up to the kinds of stuff we want to talk about, although he makes some gestures. I think he’s far stronger on the nuts and bolts of elementary emergence, along the lines of what you were talking about in the last post. This is a point of great bafflement and derision for the resisters to the complexity project. Therefore his observation that individual water molecules are not wet is a fine moment, and all of the other demos do good work. He also has things to say about how tags stack and sort, but I’m not conversational with that yet. More in a bit.

  6. To me it seems that the biggest difference between the model builders and their critics is that the model builders assume what statistician George E. Box said: “All models are wrong but some are useful.” Their critics mistake the simplification inherent in model building (a practical issue) for reductionism (a philosophical position). It isn’t.

    I recall what I read in Warren McCulloch’s introduction to Embodiments of Mind. The following is a paraphrase: I build machines that can do human things. There are always human things that my machines can’t do. The world is divided into two kinds of people, those who point to what they can’t do and say, see, machines will never be able to do what humans do. And then there are people like me. We build better machines.

    I have never read a model developer or complexity scientist who claims that her models are the final solution for any problem. Why is it that so many philosophers seek absolute certainty in this highly uncertain world?

Leave a Reply!

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: