The Science of Information Viruses

By Keith Henson

Nancho Advisory: While this treatment focuses on the pathological "viral" aspects of ideas, it might be helpful to also keep in mind their adaptive gene-like functions. Conventional wisdom implies that, like germs and Indians, the only good virus is a dead one. Yet we know that memes like liberty, democracy, small-is-beautiful, et salubrious cetera have played an actively benevolent role in human social evolution. To inoculate the earth against the abusive ideas traced below we may have to set a meme to catch (and supplant) a meme...

We don't have a science of social prediction. Until recently we haven't even had much in the way of theories. Our continual surprise at the development of cults, religions, wars, fads, and other social movements is a notable exception to the steady progress humans have made in building better models of our environment. Our lack of good models must be considered a major deficiency.

A successful theory for the development of social movements will have to provide a unifying theory for events that make up much of the evening news. It will have to discover common features that lie behind the diverse trends causing problems in Nicaragua, South Africa, Northern Ireland and the Middle East. It should be able to produce a plausible model for the breakup of the Rajneesh cult. The theory should be able to predict the conditions under which Turkey will be subverted by a fundamentalist version of Islam similar to that which has led to so much grief in Iran.

Tentative answers to these questions are beginning to emerge from the new field of memetics. Memetics (from meme, which rhymes with cream) is an outgrowth of evolutionary biology. It takes the age-old saying "ideas have a life of their own" literally, and applies models from biology to the evolution, spread and persistence of ideas (memes) in human culture.

One aspect of memetics can be thought of as "germ theory applied to ideas." Social movements can be modeled as side effects of infectious ideas that spread among people in a way mathematically identical to the way epidemic disease spreads. Drug fads, for example, have closely followed epidemic-like curves. I don't think it can be demonstrated that civil authority has any more effect on the course of the "epidemics" than it had on the course of the Black Death. At a deeper level, research in neuroscience and artificial intelligence is starting to develop an understanding of why we are susceptible to "infectious information," both the benign and the deadly.

"Meme" is a word coined in purposeful analogy to "gene" by Richard Dawkins in his 1976 book, The Selfish Gene. To understand memes, you must have a good understanding of the modern concepts of evolution, and this is a good source. In the last chapter of Dawkin's book memes were defined as replicating information patterns that use minds to get themselves copied much as a virus uses cells to get itself copied. (Dawkins credits several others for developing the concepts, especially the anthropologist F.T. Cloak.) Like genes, memes are pure information, whether the sequence is coded in DNA, printed on paper, or written on magnetic tape.

Humans are not the only creatures that pass memes about. Birds can learn variations of songs. The songs of whales are also replicating information patterns that fit the model of a meme. So is the termite-catching technique that chimps pass from generation to generation.

Meme is similar to "idea", but not all ideas are memes. A passing idea which you do not communicate to others, or one which fails to take root in others, falls short of being a meme. The important part of the "meme about memes" is that memes a re subject to adaptive evolutionary forces very similar to those that select for genes. That is, their variation is subject to selection in the environment provided by human minds, communication channels, and the vast collection of cooperating and competing memes that make up human culture. The analogy is remarkably close. For example, genes in cold viruses that cause sneezes by irritating noses spread themselves by this route to new hosts and become more common in the gene pool of a cold virus. Memes cause those they have successfully infected to spread the meme by both direct methods (proselytizing) and indirect methods (such as writing). Such memes become more common in the culture pool.

The entire topic would be academic except that there are two levels of evolution (genes and memes) involved and the memetic level is only loosely coupled to the genetic. Memes which override genetic survival, such as those which induce young Lebanese Shiites to blow themselves "into the next world" from the front seat of a truck loaded with high explosives, or induce untrained Iranians to volunteer to charge Iraqi machine guns, or the WWII Kamikaze "social movement" in Japan are all too well known. I have proposed the term "memeoid" for people whose behavior is so strongly influenced by a replicating information pattern (meme) that their survival becomes inconsequential in their own minds.

For a vivid example we can hark back a few years ago to the Reverend Jim Jones and the People's Temple incident, where 912 people, including Jones, died of complications - poison and gunshot wounds - induced by an information disease. The Children's Crusades of the middle ages were larger and more lethal; only two of 20,000 returned from one crusade. The mass suicide in the first century by the Jews at Masada is a clear example of information patterns in people's minds having more influence over their behavior than fear of death.

A more seductive example of a social movement set off by a lethal meme comes from South Africa. In the 1850's, a meme (originally derived from a dream) led to a great sacrifice by the Xhosa people during which they killed their cattle, burned their grain, and refrained from planting in the belief that doing so would cause their ancestors to come back from the dead and expel the whites. At least 20,000 and perhaps as many as 60,000 people starved when the predicted millennium of plenty failed to arrive. Known as the Cattle Killing, it was not a unique response for a primitive society being displaced by a more technically advanced one.

Memes that bring about suicidal behavior are, at least, self-limiting. Those which induce one group of people to kill another are much worse, and the social movements they induce are often much larger. The scope of the social movement known as the Inquisition is seldom mentioned in history textbooks, but:

"The number of victims claimed by the witch-hunts, which lasted for three hundred years, is reckoned by historians to be between five and six million people; it therefore caused more deaths than all the wars waged over the period... "It is only when one takes into account the brutal, pitiless expression of mass mania, and that a belief in the devil, his traffic with witches and warlocks, was constantly being fanned anew by the Church...that it is possible to gain any measure of understanding..."*
The depredations and brutality of the Inquisition were typical of deadly memes stemming from religions or closely related social movements, such as Marxist-Leninist communism. In the last decade the people of Kampuchea were infested with an anti-intellectual agrarian utopian meme clearly mutated (in the minds of Pol Pot and his close associates) from the communist meme. They were Eric Hoffer's "True Believers" of the most extreme stripe. The resulting social movement was a massive self-genocide. Over one-third of the population of Kampuchea, including almost all of the city dwellers and the educated, died before the Vietnamese (embarrassed by news stories of rivers clogged with bodies) invaded and put a stop to the killing. Many more would have died had the social movement run its course without interference. Kampuchea will take decades to recover.

History classes have made us more aware of the genocidal depredations resulting from the "master race" meme that was part of the Nazi meme complex. Considered from the viewpoint of memes, Hitler was less a prime mover than a willing victim of this particularly nasty and pervasive variety of information disease. Had plague struck Germany in the '30s instead of Nazism, we would have understood it in terms of susceptibility, vectors and disease organisms. What did happen any soon be modeled and understood in terms of the social and economic disruptions of the time increasing the number of people susceptible to fanatic beliefs, just as poor diet is known to increase the number of those susceptible to tuberculosis. Communication in the form of personal contact, the written word, radio, and amplified voices are substitutes for disease vectors. A pool of "sub-memes", many of them ancient myth, contributed to the syncretic Nazi meme in much the same way mobile genes contribute to the virulence of the influenza viruses.

Nazism was not the only fanatical movement growing and evolving in the fertile social media of Germany between the wars. The Marxist-Leninist meme was a visible competitor in the early period. Even though most of those infected with the Nazi meme were conquered or killed, and Nazism became a suppressed meme, it cannot be said to have died. As a replicating information pattern that has gone through a great deal of information honing, it still infects susceptible people today.

A fascinating footnote to the horrors of the German experience with Nazism happened in 1969 when Ron Jones, a teacher in Palo Alto, exposed a high school history class to an intensive, five-day experience with the ideas that made up the Nazi meme. The experience of that week was originally published as "Take as Directed" in the CoEvolution Quarterly (CQ #9, p.152), and a few years ago was made into a TV movie, The Wave. Over four days, Jones introduced and drilled his students in concepts of Strength through Discipline, Community, Action, and Pride. (The fifth day was devoted to showing them how easily they had started to slip into the abyss.) The enthusiasm which most of the class adopted the memes and spread them to their friends, swelling a 40 student class to 200 in five days, made it one of the most frightening events the teacher had ever experienced. Given the track record of the Nazi meme, the mini-social movement his experiment set off is no more surprising in retrospect than the medical effects would have been if the teacher had sprayed smallpox virus on the class.

An empirical characteristic of large, long-lived religious movements or related social movements (at least in the West) is a scripture or body of written material. This may function to standardize the meme involved or at least slow its evolution as the number of people infected with it grows. From Scientology right back to the Hindu Vedas, I can think of no counterexamples. Social movements involving more than a few thousand people or lasting more than a few years may have been rare before writing came along.

I have noticed several features of social movements derived from dangerous memes. One is self-isolation of the infected group, or at least of new recruits, from the rest of society. This need not be an "intelligent" action taken by the "leaders". There may be no more thought involved than the evolution of white moths into dark ones in grimy industrial England. The "fanatic cult" memes which incorporate isolation are the ones we observe; those which do not incorporate isolation are like the light moths, gone and not observable.

In the case of the Soviet Union, the communist meme survives in a society largely isolated from the rest of the world. In recent years the isolation may have resulted from reasoned considerations about the fragility of the communist meme in open competition with other memes. A more parsimonious view would note that without originally having a strong isolation component, the communist meme would have had no more social influence in the USSR than it has had in, say, France.

Isolation makes possible exposure to single meme (or meme set) many times a day for months or years without much contact with other memes. Exclusive exposure to one meme (also known as brainwashing) induces a "dependent mental state" in some people.

Thankfully, most of us have not experienced the dependent mental state firsthand, but we have all seen such people in the news programs boarding buses for the front in Iran, or been harassed by them in airports, or had them knock on our doors and try to infect us. It is clear that people who suffer from extreme cases of "information disease" have lost much of their ability to take care of themselves or their children. Truly dedicated people often fail to replace themselves, since too much of their life energies are channeled into propagating the infecting meme. One example comes from the largest sub-division of Christianity, where celibacy among its most dedicated has long been institutionalized. The Rajneesh cult practiced the opposite of celibacy, but discouraged births to the point of sterilizing the barely pubescent female children of its resident members.

Given that memes have been interfering with our reproduction for a long time, one must wonder why humans are still so susceptible to information diseases. The answers to such questions are starting to come from research into artificial intelligence, neuroscience and archeology. It is becoming apparent that our vulnerabilities are a direct consequence of the way our minds are organized, and that organization is a direct consequence of our evolutionary history.

Marvin Minsky (a principal founder of AI) and Michael Gazzaniga (one of the major workers in split brain research) have independently come to a virtually identical model of the mind. Both view minds as vast collections of interacting, largely parallel (co-conscious) modules or "agents", or a "Society of Mind."** The lowest level of such a society of agents consists of a small number of nerve cells that innervate a section of muscle. A few of the higher-level modules have been isolated in clever experiments by Gazzaniga, some of them on patients whose right and left hemispheres have been divided by trauma or surgery.

One surprise from this work is that we seem to have our mental modules arranged in a way that guarantees that we will form beliefs. What we believe in depends, at least in part, on what we are exposed to and the order in which we are exposed. Gazzaniga argues that we slowly evolved the ability to form beliefs because the ability provides a major advantage in surviving. Being able to infer, that is to form new beliefs, and to learn, in the sense of acquiring such beliefs from others, was a major advance over learning by trial and error. Being able to pass the rare new ways of our ancestors found for chipping rock or making pots from person to person and generation to generation was vital in allowing humans to spread over the Earth.

But as this ability became the norm, communicating human minds formed a "primal soup" in which a new kind of non-biological evolution, that of replicating information patterns or memes, could get started. A wide variety of competing memes has evolved in the intervening seventy thousand years or so. It should not be surprising that the survivors of this process, like astrology or religions, are so effective at inducing their hosts to spread and defend them. It is also plausible that in the tens of millennia since memetic evolution became a major factor there has been biological co-evolution. The parts of our brains that hold our belief systems have probably undergone biological adaption to be better at detecting dangerous memes and more skeptical about memes that result in death and seriously interfere with reproductive success.

This type of co-evolution is known as an "arms race" to biologists. One such biological arms race has resulted in almost perfect egg mimicry by the cuckoo and in correspondingly sharp visual discrimination in the birds it parasitizes. By analogy, while we get better at spotting dangerous memes, the memes may be evolving to be more effective at infecting us. Advancing technology (which itself is an improving collection of memes) changes the environmental conditions where memes survive or fail as well. The modern telephone system and the tape cassette player were major factors in the takeover of Iran. It has been argued that the rise of the Nazis depended strongly on radio reaching a previously unexposed and unsophisticated population.

I have picked dangerous examples for vivid illustrations and to point out that memes have a life of their own. The ones that kill their hosts make this hard to ignore. However, most memes, like most microorganisms, are either helpful or at least harmless. Some memes may even provide a certain amount of defense from the very harmful ones. It is the natural progression of parasites to become helpful symbiotes, and the first such behavior that emerges in a proto-symbiote is for it to start protecting its host from other parasites. I have come to appreciate the common religions in this light. Even if they were harmful when they started, the ones that survive over generations evolve and do not cause too much damage to their hosts. Calvin (who had dozens of people executed over theological disputes) would hardly recognize Presbyterians three hundred years later. Contrariwise, the Shaker meme is now confined to books, and the Shakers are gone. It is clearly safer to believe in a well-aged religion than to be susceptible to a potentially fatal cult.

History doesn't change, but our interpretation of it can. For example, the contemporary "causes" of historical epidemics (such as the miasma theory) have been totally supplanted by germ theory explanations. Before germ theory came along, memes of causality for epidemics were remarkably stable. The "explanation" for the Black Death of l348 was still in use for the Philadelphia Yellow Fever epidemic of 1796. Similarly, various "explanations" for wars have been with us for hundreds of years.

Memetics provides a new way to analyze recent wars and the roots of current disputes. In this view, the ultimate (though unaware) protagonists of World War II were memes such as the Nazi "master race," and the Marxist-Leninist meme (MLM). The current clash between the Soviets and the Western world can be viewed (in a grossly simplified way) as a meme conflict for space in minds between the religion-like, competition-intolerant mono-meme of communism and the western tolerance meta-meme. While it is not a religion by any reasonable definition, the Marxist-Leninist meme is clearly in competition for the "belief space" in minds usually occupied by religious memes. It, and its more cultish offshoots, have the typical virtues and excesses of cult-stage religious memes. In an amusing twist, the "godless" communist meme is the more religious in tenor of the two in its battle for mind space with secular Western culture!

Reviewers of an earlier draft of this article objected to my description of Soviet memes. Words like "tolerant" and "intolerant" have acquired a great deal of positive/negative connotation in the Western world, but in describing memes, I am using them in the same way we would say that a mold colony is intolerant of a bacterial invasion. With respect to the belief system that dominates the meme pool of the other superpower, I am trying to be descriptive, not partisan.

If anything, i would think that understanding the memetic nature of religions and related movements like communism would defuse the emotional connections and substitute something closer to dispassionate understanding of the parasitic-to-symbiotic memes behind such social movements. It has had that effect on me. Even the most gruesome features of communism are what they are simply because those features were (and are) necessary for the meme to exist in a world of competing memes. Isolation, for example, is a common feature of virtually all successful religious-type memes while they are in the cult stage. Anyone who has studied history knows that suppression of competitive memes by the power of the state is a common experience once a meme of this class has infected the leaders or they have been replaced by those infected. And if the Christian religion was a mainstay of the aristocracy, serving to keep the peasants in place, Soviet Communism is no less supportive of its own hereditary elite. As a successful and persistent meme, that has appeal even to people who know the realities of its practice. It commands a certain grudging respect.

From a meme's viewpoint, tolerance of other memes is not a virtue. It is, in fact, a fatal characteristic for a particular meme, as memes inducing intolerance to other memes would soon displace it. On the other hand, a meta-meme of limited toleration, or even cooperation among memes, is possible. The Western meta-meme of tolerance seems to have emerged from an ecosystem of memes in much the same way that cooperative behavior has been modeled as emerging from an ecosystem of individuals.*** In the area of meme tolerance the Western world may be unique. We think of censorship as evil; where but in an advanced ecosystem of memes could such a strange idea have emerged?

There are historical traces for the development of the meta-meme of tolerance. This particular character of our ecosystem of memes has been developing at least since the writings of the Greeks and Romans were rediscovered during the Renaissance. Studying inactive pagan religions may have been the first step in developing a tolerance for a variety of religious memes. The fragmentation of the dominant religion during the Reformation led to a series of largely indecisive religious wars in most of the major countries of Europe. Sheer exhaustion may have been one of the most significant factors in developing a grudging tolerance, which in these later times has taken a patina of virtue in the division of our culture known as "liberal."

In this view, Western culture is a vast ecosystem where memes of many classes engage in "fair" competition with each other. Attempts to subvert fair competition by changing laws or education (such as introducing "creation science" into schools) draw opposition from defenders of a wide variety of memes which have evolved within this environment. This model may provide testable explanations for both Western culture's tolerance of intolerant memes (such as creation science and the MLM) and the hostility these memes evoke from various segments of the culture. David Brin's "Dogma of Otherness" in the April l986 issue of Analog is recommended for those who want to consider the origins of such peculiar ambiguities in our culture.

Several current social movements are obvious candidates for examination with memetic theory. Given the available data, we may be able to predict the remaining course of the "non-literate graffiti epidemic," which has spread in the past 15 years from New York City to remote corners of the country. There are substantial financial reasons (such as the cost of mark-resistant walls) to want to know if scribbler behavior will be a limited epidemic, or will this behavior become an endemic part of our culture?

Drug use, clearly a replicating pattern of behavior passed from person to person, is another "social movement" where the similarity to epidemic waxing and waning has been widely used by reporters, and noted without much explanation in a number of learned journals. If it were formally considered as an epidemic with memes as the infecting agents, the ways by which the behavior spreads might get more attention. Antidrug programs might be evaluated in terms of how well they induce reasonable behavior. Some efforts in the past, especially those which wildly exaggerated the dangers of a drug such as marijuana, may have increased the behavior of taking other drugs. These efforts may have immunized those exposed against believing any official pronouncements about drugs.

Formal consideration of drug use as an epidemic of meme-induced behavior might also lead to the realization that the percentage of people susceptible to abusing most drugs is not all that large. (Cigarette smoking is an exception.) For example, most of the people I know who have tried cocaine don't care for it. Not liking the effect, they wouldn't use it if it were free. People who really like opiates aren't that common either.

If most conflict in the world is an indirect effect of memes, memetics holds as much potential for reducing human misery as the germ theory of disease. Just being able to model the interactions among the Soviets, the West, and the Islamic groups may make the world a safer place. Widespread understanding of hard-to-avoid human susceptibilities and an ecosystem-like model of replicating information patterns that have no short-term interest in their host (and indeed no consciousness at all) may lead to the development of meme evaluating "mental health practices" just as knowledge of disease has changed our behavior in regard to drinking ditchwater.

If this article has succeed in infecting you with the meme-about-memes, perhaps it will help you be more responsible about the memes you spread and less likely to be infected with a meme that will harm you or those around you.

* p.163, Five Thousand Years of Medicine, Gerhard Venzmer; Taplinger Publishing, NY '68.
** The Society of Mind, Marvin Minsky; Doubleday, NY '86.
*** The Evolution of Cooperation, Robert Axelrod; Basic Books, NY '84.

This article reproduced from the meme-intensive
"Signal" issue of Whole Earth Review, #57

Memework | Main Archive