Swarm Theory and Politics

First published at Opinion Forum.

Can analogies be drawn between insect swarms and human political activity?

Stink bugs have been swarming in Washington, D.C. where, unlike CongressCritters, "they really cause no harm. . . . just a nuisance." Fortunately, the Congress has adjourned.

The swarming addressed here can be more than just a nuisance. Sometimes the results can be seen as good or bad depending on what one wants. Swarming (as distinguished from smarming, a human social and political tactic) is a natural phenomenon permitting essentially brainless creatures -- ants, bees, other insects and some fish -- to thrive through apparently instinctive collective reactions to external stimuli. Maybe swarming also illuminates some species of political behavior.

Ants and bees construct truly ingenious structures in which to live and breed and which they protect when endangered. If a bee hive is destroyed the surviving queen leaves, followed by a swarm, and they get busy building a new hive -- without the help of FEMA, federal grants or other financial stimuli. Ants do much the same with great industry. The nests of some ant colonies extend for many miles underground.

In Europe, one vast colony of Argentine ants is thought to stretch for 6,000km (3,700 miles) along the Mediterranean coast, while another in the US, known as the "Californian large," extends over 900km (560 miles) along the coast of California. A third huge colony exists on the west coast of Japan.

As with bees, ant colonies are marvels of architectural sophistication. Highly unintelligent creatures, apparently lacking even a sense of their own existence or that of their fellows, acting collectively and instinctively in swarms, they can do some things as least as well as can individual humans and can exhibit collectively something resembling individual human intelligence and activity. As a point of at least minor interest, humans are thought to have a biomass of approximately 335 million metric tons and ants of between 900 and 9,000 million metric tons. Bee biomass is thought to be declining, but I couldn't find any numbers.

When an ant leaves its nest to forage for food it takes an apparently random path. When it finds food, it carries it back to the nest leaving a trail of chemical signals called pheromones. Other ants also go out looking for food and when they find it they do the same things. There being no welfare system, when the first ant returns with food others follow its pheromone trail back to the food source and return by the same path, leaving additional pheromones. The pheromones are short lived, so once a food source is exhausted and unsuccessful ants have abandoned it and gone looking for other sources, the old trails evaporate. When new sources of food are found the ants returning to the nest with food leave new pheromone trails to be followed by others until they in turn evaporate. There is a lot more to it than that and there is a mountain of literature on the subject of swarming; a bit of it is available here.

Computer programs based on ant behavior abound and parcel delivery services use them daily to determine the most efficient routes for their trucks. Much of the traffic on the internet is routed using similar ant-based algorithms. Some suggest that the human brain itself involves swarm theory, "even complex cognitive functions, such as abstract reasoning and consciousness." Maybe our brains resemble bee hives, although in some cases that might be an insult to the bees.

In 2002, Michael Chichton (1942 - 2008) wrote Prey about a nanoparticle project gone very wrong. Nanoparticles are microscopically small, dust like things. The idea was to use swarms of nanoparticles, some with visual, some with aural and some with other capabilities, to conduct military surveillance and for other purposes. However, they inexplicably "evolved" very quickly -- a generation was of about three hours duration -- and began to reproduce as well as to swarm in unforeseen ways. The novel leaves unanswered the question of whether the human race survived. Scary and far fetched, but not as far fetched as one might think. The United States military have been looking seriously at swarm behavior since before 2000. Murphy's law, that whatever can go wrong will at the most inopportune time, is similar to the concept of the fog of war, where ambiguity, chance and chaos, with all of their non-linear recursive complexities, continue to play important roles despite (or perhaps because of) advances in command and control equipment and procedures leading to information overload. Small events can have large consequences. Multiple articles are collected here. The military apparently envision using large numbers of inexpensive, highly expendable and essentially autonomous robotic devices to accomplish various missions. These are not the remotely controlled unmanned drone aircraft currently in the news.

[C]urrent UAVs require at least one operator per UAV, despite technological advances that make it possible to deploy hundred (if not thousands) of inexpensive [UAVs]. This requirement not only increases expense, but makes coordination among UAVs more difficult.

Recently swarm technology has been suggested as a possible solution to automatically control and coordinate multiple UAVs. Swarms consist of a large number of distributed, parallel-acting individual entities coupled with primitive communication mechanisms such as chemical markers. The idea behind a swarm is that simple local rules that govern the behavior of individual entities can lead to complex emergent [self-taught] behavior of the system as a whole. For example, it might be possible to conduct a search and destroy operation. Rules include ideas such as “avoid areas already searched” or “avoid UAVs within a certain radius.”

Cheap and stupid robots are in some cases more effective than teams of intelligent soldiers with assigned specialties. Soldiers think and adapt quickly to changing conditions. The loss of even one with a critical task can cause a mission to fail. Although robots don't "think," they too can react very quickly to changing stimuli. They are homogeneous and can be numerous. The loss of some is matter of little consequence to the mission. Are nano devices being considered? So it appears.

People obviously (with few exceptions) are individually far more intelligent than are ants and human swarms are less common than with bees. Still, sometimes they occur. At a hotly contested rugby match, something similar to a swarm sometimes occurs. A sense that team A won unfairly (or simply because it was team A rather than team B) can cause a major riot of thousands of spectators. "Deprived youths" in some nascent Islamic countries swarm to offer burnt automotive sacrifices to their image-phobic god, while others, perhaps more devout, riot self-destructively to allay cartoonish offenses to theirs. A racial incident can cause similar rioting, particularly if stimulated by queen bees and lesser luminaries; great and self-destructive damage to the infrastructures of the communities from which the rioters hail is common. They "lose their minds" and become a swarm. "Damn those jerks! We'll show them. Let's burn our houses! Yeah! Let's!" The phrases "herd mentality" and "mob mentality" are commonly applied. This sort of thing occasionally happens in politics. The politically inspired race to judgment over the activities of the "elite" Duke University lacrosse team vis a vis a stripper may have been a minor example of something similar; the queen bees were very busy.

On the other hand and in a different context, some beneficial things can result from conduct resembling but not amounting to true swarm behavior.

[T]op-down management is viewed as a forced, and thus, ineffective approach. We say that the bottom-up thrust is based on the world's creatures' success with 'emergence', 'swarm theory' or 'chaos theory' a bottom-up approach used by ants, bees,birds and even trees[??]. Its basis is that small simple local decisions, if allowed to develop and follow their own path under an overall strategic guidance of process, will lead to an elegant solution - not just a better solution but a superior solution that no single individual could likely have conceived.

I don't know about the trees, but does this at least remotely suggest T.E.A. party organizations to anyone else? According to this Wall Street Journal article, individual groups called "tea parties" remain highly diverse but are coalescing in pragmatic fashion to the extent necessary to achieve their basic goals. It seems to be a bottom up, and far from top down, phenomenon.

Statism and the accompanying central (top down) planning, direction and control of human conduct are fatally flawed but might work well in an ant colony or bee hive, provided the insects were not thereby deprived of their autonomous nature. If Venezuelan el Presidente Chávez could wiggle his buttocks like a queen bee and invoke consistent and reliably productive responses by the masses, Venezuela might be a viable machine for creating stuff. He can't and it isn't because that's not the way in which humans have evolved. There are other problems:

we know that complex systems are exquisitely sensitive to initial operating conditions and that interventions can have disproportionately large or small implications depending upon timing, etc. . . .

Central planning does not, and cannot, take into account apparently insignificant changes which can have very important effects.

North Korea seems to offer the best current example of remarkably unsuccessful quasi-swarm-based human endeavor. Fortunately, elsewhere in politics there tends to be substantial diversity of opinion transcending the charisma of queen bees. Although the power of a queen bee once firmly entrenched is difficult to undermine it is possible. Senator Obama managed to displace Senator Clinton as queen bee in 2008. Now many of his swarms of worker bees are wondering what happened and may be looking for a replacement. Many who didn't like but voted for him in 2008 are also doing so. Does Governor Palin seek to be Queen Bee (at least for a day)? I don't think so and hope not.

Humans are neither ants nor bees and we don't have the same instinctive reactions. Leaders are necessary and as long as their pheromone trails lead to what we desire we follow them. Also like ants, when they don't we stop. Unlike ants and the folks in North Korea, we get to decide what we want and whether the paths directed by our leaders are likely to be fruitful. Sometimes we screw up in deciding what we want and how to achieve it, and we ultimately pay the price as we -- and the queen bees -- must. When members of the political class see the "unwashed masses" as bees ready to swarm, they sometimes are not too far off and sometimes they are. Even bees and ants respond to new external stimuli and their behavior can't always be predicted accurately; the results can be chaotic and chaos theory is very much implicated in swarm theory; maybe with an explanation by Uncle Jay it would all make more sense.

Does swarm theory have any significant relationship, at least by analogy, to political behavior or would a suggestion that it may be faddish? I don't know, but the concept is worth keeping in mind as we decided how to cast our votes next month and reflect upon the outcome.