Is group selection coming back?

Elena Kleshchenko, "Chemistry and Life" No. 5-2008

In the winter of 2007-2008, a "battle of titans" took place on the pages of English–language scientific journals and on the Internet – two prominent figures of modern biology, two famous popularizers of science, and in their person - the English and American evolutionary schools. The first of them, Richard Dawkins, is well known to our readers: the author of the concept of the "selfish gene", who proclaimed the gene as the elementary unit of selection; the one who introduced the term "meme" to denote the unit of information replicated in our minds; a passionate Darwinist and no less passionate anticlerical, the author of books about science that instantly become bestsellers- this is not a complete list of his achievements (for more information, see the official website of Dawkins). Dawkins' arguments are strong, his fame is huge, and his meekness and tolerance for opponents are proverbial, so not everyone will risk publicly objecting to Dawkins on his field.

However, Edward Wilson is "not everyone". Although this American biologist is much less famous in Russia – he is only the founder of sociobiology, the science of the biological foundations of social behavior, and the winner of two Pulitzer Prizes: 1979, which brought him the book "On Human Nature" ("On Human Nature"), and 1991, for the book "The Ants" – "Ants". Wilson is not one of those sociologists who talk about biology without knowing its basics – he is a major expert on ants and other social insects. It was he who publicly stated that behavior is largely determined by genes, not only in insects, but also in humans, and, speaking about this, even used the term "genetic leash". Such expressions offended many: but what about, they say, freedom of will and education in the spirit of democratic (variant: communist) ideals? Wilson has been called both a racist and a eugenicist, although he is no more a racist than academician E. D. Sverdlov (see "Chemistry and Life", 2008, No. 3) or the same Dawkins: he can rather be called a representative of scientific humanism. The purpose of the book "On Human Nature" was "to complete the revolution started by Darwin and bring the achievements of biological thought to the social and humanitarian sciences." Nevertheless, many ethologists, including E. N. Panov, who is familiar to our readers, are critical of Wilson's ideas. However, Dawkins himself speaks of Wilson and his writings with unwavering respect, even in the heat of an argument.

So where did the author of the "Selfish Gene" and the father of sociobiology disagree, both Darwinists, educators and anti-prejudice fighters? The apple of contention was the group selection.

Species or gene?

In Dawkins' books, "group selection" is actually an expletive. There is no group selection, and all the talk about it is a harmful delusion or mental laziness. "A group such as a species or a population within a species whose individual members are willing to sacrifice themselves for the well-being of this group has a better chance of avoiding extinction than a rival group whose individual members put their own selfish interests first. Therefore, the world turns out to be populated mainly by groups consisting of selfless individuals. This is the essence of the theory of "group selection", which biologists who are not familiar enough with evolutionary theory have long considered correct... The orthodox alternative theory is usually called "individual selection," although personally I prefer to talk about gene selection." Among the biologists "insufficiently familiar with evolutionary theory," Dawkins, not without regret, includes Conrad Lorenz.

Let us briefly recall the arguments with which Dawkins proves that in fact the unit of selection is a gene, and not an organism, and even more so not a species or population. As for the species – indeed, it is unlikely that an animal can show "concern for the welfare of the species as a whole", die for its prosperity. Why would a Moscow sparrow care, for example, about the sparrows of Vladivostok, if he does not even suspect their existence?

Dawkins also questions the term "individual selection". It does not seem that the main goal of evolution was the prosperity of the strongest individuals. After all, the winner of the evolutionary race, instead of enjoying the fruits of victory, starts breeding at the first opportunity, spending a lot of effort on finding and winning a partner, on the "process itself", on caring for offspring (often sacrificing not only prosperity, but health and life) – and in the end inevitably grows old and dies. What kind of prosperity is there.

The only ones who win in this game are the genes of individuals who managed to continue themselves in descendants. They create for themselves "survival machines"– the bodies of animals in whose cells they live– in much the same way as the first nucleic acid molecules built the first protocell. By ruthlessly programming the behavior of "survival machines", genes encourage them to engage in self-copying (forgetting about food and sleep: getting new copies of genes is more important), to look for partners (the best, because genes want to be in good company when they move into the bodies of offspring), to take care of cubs (about new "survival machines" containing copies of the same genes) – and it is possible that it is the genes that command the old bodies to leave the stage, freeing up space. However, according to Dawkins, genes are rather indifferent to the fate of old bodies from the moment the offspring grew up. If there are enough resources, let them live, suddenly grandma will help nurse a new series of copies, and grandpa will take a young wife and start copying again...

In all his books, starting with The Selfish Gene (1976, Russian translation – 1993), Dawkins tirelessly emphasizes two things. Firstly, expressions like "genes are ruthless" or "indifferent" should be understood as a metaphor: genes do not have any plans and do not experience any feelings, because parts of the DNA molecule are not capable of this in principle – they simply participate in certain chemical reactions. Secondly, the selfish behavior of our genes is in no way an example to follow: that's why we are the kings of nature, so as not to reach for the "gene leash".

Among biologists, the clear, coherent and consistent theory of the "selfish gene" quickly gained popularity. After all, behind these arguments there was a huge work of many modern evolutionists: field observations, the results of laboratory experiments and, finally, mathematical and computer models (the first of which were performed on "large calculating machines" of the 60s, but have not ceased to be convincing to this day). However, all this did not prevent the opponents of the "selfish gene" from accusing Dawkins of immorality and reductionism. We will not consider the first accusation, but what is there with reductionism?

The green beard is a symbol of altruism

If group selection does not exist, they asked, then how does the theory of gene selection explain the observed facts of altruism in animals? What is the benefit for the gene if the "survival machine" dies to save a relative? This theory has no problems with the self-sacrifice of a parent in the name of children. There is no altruism as such, but there is kin selection (from kin - relative): the parent's gene saves its own copies in children, without ceasing to be selfish. In other words, all animals whose offspring will not survive without care must necessarily have a gene complex that ensures love for the cubs: for them, the sight of happy and prosperous descendants is a reward, or, as experts say, positive reinforcement, but an unhappy descendant is a terrible test for nerves. Irresponsible moms and dads, whose parental love genes do not work properly, do not transfer damaged copies to a new generation, since their offspring are likely to die.

Kin selection also explains the brotherly or sisterly care, usually of the cubs from the previous litter – about the younger ones: if the cubs have both mother and father in common, then they have half of the genes in common. It is said that when John Haldane was asked in a pub if he was ready to die for his brother, the famous evolutionist quickly calculated something on a piece of paper and then answered: "Only for two brothers or at least eight cousins!" is, of course, a joke, but it shows how kin selection works: the act of altruism is encouraged by evolution if it leads to the preservation of more copies of the gene than selfish behavior.

The mathematical expression of this concept was proposed by William Hamilton in 1964. In its simplest form , the Hamiltonian rule looks like this:

    rb > c,

where r is the degree of kinship between the altruist and the object of altruism, b is the benefit for the relative that the altruist's help will give, and c is the damage that he will suffer because of his kindness (from the English words relatedness, benefit and cost). When this inequality is respected, an act of altruism is possible. For example, it is very possible to remove a biting parasite from the back of a neighbor's head with an excess of free time (c – close to zero), but it makes sense to rush into a stormy river only if your cubs are drowning, and you somehow know how to swim. If you swim well, you can save your nephew. But if you don't know how at all, it's more practical to have new cubs. Of course, in order for such a model to have predictive power, it is necessary to express in numbers not only the degree of kinship, but also damage with benefit, for example by assigning them points.

A reasonable question: and how do animals (even smart enough to distinguish relatives from each other) find out who their relatives are? After all, they don't have surnames and passports! Let's say mothers know their cubs, but fathers have a more complicated situation, and how do brothers know if they are relatives or stepbrothers... It can be answered that this is why maternal care in birds and mammals is more common than paternal and even more fraternal. And besides, it is shown that many animals distinguish their relatives very well, for example, by smell. This applies to both social insects and mammals. True, mistakes are possible here if a genetically "friend" has spent a long time among "strangers" – up to the most fatal consequences.

In the end, you can just focus on the appearance. Hamilton predicted the existence of the "effect of altruism to a green beard" (the name was suggested by Dawkins): if, for example, some individuals in the population have a green beard, it is possible that green-bearded individuals will be more willing to help each other than, say, red-bearded ones. Gene egoism justifies such behavior: if not all greenbeards are close relatives, then the genes responsible for the color and presence of a beard are probably the same. But it is not so easy to detect this effect in its pure form: after all, it is necessary to prove that they help not a relative, but a similar-looking creature, moreover, more often than a dissimilar one, and not in exchange for reciprocal help (this will be discussed later), namely for similarity...

The "green beard" turned out to be a blue chin. Five years ago, New Scientist wrote about the spotted lizards Uta stansburiana. In males of these lizards, the neck can be colored blue, orange or yellow. It turned out that males with a blue neck enter into cooperation for the protection of the territory with other blue males, and under certain conditions this defensive alliance may be beneficial not for both, but only for one of the males: one mates with a female, and the other "stands on guard" and drives away competitors. The most genuine altruism! The authors of the work conducted at the University of California (Santa Cruz) under the leadership of Barry Sinervo believe that this is the first indisputable example of the "green beard effect" in vertebrates. Interestingly, males with chins of other colors are not inclined to cooperate: "orange beards" are aggressive, and "yellow" prefer to sneak into someone else's territory to someone else's girl. Now researchers have established which genes are responsible for this behavior of "blue". To everyone's surprise, it turned out that the genes controlling the color of the chin and at least three other genes determining behavior are not linked, but are located in different parts of the genome.

It is difficult to resist the temptation to use the idea of "altruism to the green beard" to explain human relationships. After all, it has been noticed more than once how redheads, bespectacled people and those who read the same books in childhood sympathize with each other. (We emphasize that we are not talking about marital selectivity, but about friendly sympathy.) And if a little more serious – who knows if the same effect lies at the heart of the racial prejudices of our species?

A fang for a fang, a tail and a half for a tail...One way or another, when a cub drowns, an adult has no opportunity to smell it or examine it.

Apparently, this is why the innate program of many primates involves helping any animal that looks like a cub of their species – it does not matter whether it is a native, a second cousin, or even belonging to a neighboring pack. Sometimes such help can also cross species boundaries – do you remember the stories about dolphins rescuing people, or hares adopted by a dog? The stories are very touching, but, according to sober-minded biologists, there is still an erroneous triggering of the program. Moreover, all of the above does not negate the fact that a lion can devour the cubs of a beloved lioness from another male, so that his children are born sooner. And even among people, the frequency of murders by the husbands of the mother of young children from another marriage is many times higher than similar statistics for native fathers...

However, in addition to altruism to relatives, there is also reciprocal altruism ("you are to me, I am to you"): you can take a risk for the sake of a relative, if he then takes a risk for me. And if we take into account that a single animal can pay good for good, or provide services to all relatives in any case, or cheekily use someone else's kindness, or act differently in different cases, then it becomes clear that one formula cannot do here. To describe such difficulties, J. Maynard Smith proposed using mathematical models of evolutionary strategies, again awarding points to an animal for help received from another individual, and subtracting – for the work spent on helping others, or for being left without help. Such models show what will happen to a group consisting of egoists alone, or of highly moral types who always help their neighbor (these will live perfectly – until an egoist appears among them, for example, as a result of mutation), or of those who follow the principle "as with me, so am I", or from individuals with different levels of altruism... Maynard Smith coined the term "evolutionarily stable strategy" (ESS) – such a strategy that is not displaced by others and does not lead to the death of the population, but contributes to its prosperity.

Several different strategies can coexist in a population in equilibrium. A perfect example is the same spotted lizards, whose pugnacious "orange" males, cunning "yellow" and "blue" inclined to camaraderie, according to Sinervo, endlessly play "rock–paper-scissors": "orange" beat "blue", "blue" - "yellow", "yellow" – "orange", as a result, the number of each variety varies from season to season, and none wins completely.

By the way, points are points, but in real life, altruistic behavior does not always benefit even the object of altruism. E. N. Panov in his book "Escape from loneliness" told about the observations of the Israeli ornithologist Amotz Zahavi for talkers. In these birds, older children who do not have their own territory can stay with their parents and help them raise chicks, but this help is accompanied by "communal squabbles" and often leads to the death of eggs (see Chemistry and Life, 2003, No. 5). However, in other bird species, relations in the communes are well established, and the help really increases the survival rate of the offspring.

One way or another, according to proponents of gene selection, the spread of altruism in the population is solved at the level of genes. The contribution of group selection to this process, if there is, is negligible, and the success of groups consisting of altruists, if observed (and the opposite may be observed, let's recall the talkers) – a consequence of genetic selection.

Multilevel selection

Edward Wilson co-authored an article rehabilitating group selection entitled "Rethinking the Theoretical Foundations of Sociobiology" with his namesake David Sloan Wilson from the University of Birmingham. It was published in the journal "Quarterly review of biology" (December 2007, vol. 82, No. 4), and an abridged version – in "New Scientist". The key question of the article: does it not seem ridiculous to the scientific world that we have to explain the signs that increase the fitness of a group in comparison with other groups by the action of intra-group selection, although within the group altruism is rather harmful for the individual? And isn't it because talking about intergroup selection has become unfashionable and dangerous for scientific reputation for some time?

Charles Darwin also wrote in The Origin of Man: "Although high moral standards give little or no advantage to the individual and his offspring in comparison with other people of the same tribe... the development of high moral standards undoubtedly gives a huge advantage to one tribe over another." Darwin, on the other hand, explains this amazing fact – the spread of altruistic traits in the population that are unprofitable "in private life" – by the action of group selection: altruistic groups gained advantages over groups consisting of egoists and became more numerous, thereby spreading the altruism gene. By the way, in the field of E. Wilson's professional interests, this conclusion is confirmed: colonial hymenoptera account for two percent of the total biodiversity of insects in the tropical forest, but if we count by biomass, then social insects account for three quarters. Solitary species and primitive colonies do not stand up to competition, just as other primate species cannot seriously compete with humans.

However, in the 50s and 60s, the theory of genetic selection prevailed: J. Haldane, J. Maynard Smith, W. Hamilton, and finally Dawkins as a scientist and popularizer and his followers. Computer modeling and the disclosure of the molecular nature of genes certainly played a role. Evolutionists seemed to have the keys to wildlife in their hands, and indeed the new approaches proved fruitful. However, intergroup selection was less amenable to modeling (simply because of the complexity: it was unrealistic to approach such a task while "machine time" remained a rare and expensive resource). And in any case, his contribution seemed negligible: the fluctuation in the number of groups is a slow process, their influence on each other is complex and ambiguous, because in addition to food and territorial competition, there is also, for example, the exchange of marriage partners... Is it worth taking this mythical group selection seriously if the spread of successful adaptation is perfectly explained in terms of gene selection?

The Wilsons' answer: yes, it is, at least when it comes to the genes of social interactions. After all, these genes, by definition, manifest themselves only at the group level. (Recall that for ethologists and mathematicians, even two individuals are already a group, and a family of five individuals is a complex system.) Although the unit of natural selection is certainly a gene (it would be strange to think otherwise 50 years after Watson and Crick), both an individual and a social group can be the target of selection, depending on which trait selection acts on. Therefore, it makes sense to talk about multilevel selection (D.S. Wilson's term): some genes are distributed in the population by selection at the level of individuals, and others by selection at the level of groups. Thus, the general direction of the evolution of social species is defined as the sum of two vectors – certainly multidirectional, and possibly more than two if there are several levels of socialization.

A small step from the private to the general

In the 60s, it was assumed that there was no convincing evidence of the existence of group selection. (And those observations that seemed to point to it were interpreted differently, at least through kin selection.) Now such evidence has been obtained. True, so far on microorganisms, but many discoveries have been made on them.

The bacterium Pseudomonas fluorescens has an adaptation to an oxygen-poor environment: bacterial cells secrete cellulose, which floats up and forms a mat at the surface, then attach to it and "breathe" for their pleasure. In cell culture, there can be both conscientious manufacturers of mats and freeloaders who attach themselves to other people's mats. The freeloader's strategy is clearly beneficial: the cell can not waste energy on the synthesis of life-saving means, but calmly eat and share. However, their number in culture never increases excessively. And the reason is simple: mats overloaded with parasite cells are drowning.

By the way, Dawkins wrote in the comments to the next edition of the Selfish Gene: "Perhaps there is some higher-level selection,"intereso-selection, favoring reciprocal altruism. On this basis, it is possible to develop an argument in favor of a kind of group selection, which, unlike most theories of group selection, could be acceptable." He, however, reacted extremely sharply to the slip in the Wilsons' article that, they say, Dawkins admitted his mistakes regarding group selection. He wrote an open letter to the Wilsons, which he published on his website and in the New Scientist, and demanded an apology: "I, as a scientist should, always rejoice when something makes me change my mind, but this is not the case."

Perhaps the reader is already confused: what didn't they share? Wilson and Wilson recognize the role of genes and kin selection (and do not forget to mention that the cradle of almost any animal association was a group of closest relatives), Dawkins does not object to group otb... that is, we apologize, the selection of a higher level. So what is the dispute about, other than terminology? Because of the driving force of the process.

Edward Wilson introduces the concept of major transition – let's translate it as a "qualitative leap", because the word "transition" is already occupied by geneticists. All groups of animals that gave rise to genuine social associations experience an important change at some point: competition between individuals weakens, giving way to altruistic tendencies, and from that moment intergroup selection begins to operate. It is in this way that solitary hymenoptera, wasps and ants move on to social life. There are known "intermediates" of this process, for example, a bee from the Formosa Braunsapis sauteriella. In this species of bees, the uterus lays eggs in the hollow stems of plants – a primitive analogue of the hive, but the larvae are already taken care of by worker bees who have "refused" reproduction. "One small step of a bee, a big step of all hymenoptera," comments E. Wilson.

And since this step has been taken, according to Wilson, for social animals, Hamilton's formula should be clarified:

    rb + b _e > c,

where b _e – the benefit for the group as a whole – can be many times higher than the first term of inequality, "allowing" acts of altruism that were previously impossible.

The transition to altruistic behavior can be caused phenotypically (only nutrition determines whether a female larva will become a worker bee or a queen), but it can also occur at the gene level. There is a known species of ants imported to the USA from South America, which literally before the eyes of scientists moved from a "family" lifestyle with one or two queens to life in colonies with a large number of unrelated queens, and at the same time the ants stopped defending the territory from enemies. The cause was a mutation in one gene responsible for recognizing odors.

It is noteworthy that the "qualitative leap" according to Wilson should have taken place both at the beginning of the history of life on Earth (namely, with the appearance of multicellular), and at its end – with the formation of intelligent man as a species.

The first multicellular and the first humans

It is obvious that cells, uniting into a colony, must move from competition to cooperation: competition between cells of one organism, the "selfish" reproduction of individual cells to the detriment of the whole does not lead to anything good. But this does not mean that the "qualitative leap" that led to multicellularity is necessarily accompanied by the extermination of all egoists in the colony.

New Zealand microbiologist Paul Rainey, who studied Pseudomonas fluorescens, argues that the prototype of the first multicellular can be seen precisely in the bacterial mate, which is populated by workers synthesizing cellulose for the needs of the colony, and idlers engaged only in reproduction (Paul V. Rainey, "Nature", 2007, vol. 446, p. 616). After all, if you think about it, reproduction in nature is not idleness, but the most important thing. And cells that are not engaged in the construction of the mat (which means that they are attached to it less firmly) and have saved all the forces for division could take on the role of germ cells – to produce cells that will not remain on the mat, but will float away to give rise to new colonies. And the builders of the mat will play the role of somatic cells that ensure the growth of the colony's "body" and the prosperity of germ cells. But at the same time, of course, the difference between "somatic" and "sexual" cells should become non-inherited: genetically they should be the same, and differences in "behavior" will relate to the sphere of modification variability. (Let's say a cell that finds itself in a solo swim will be programmed in such a way that when dividing, one of its halves will certainly start building a mat, like all its descendants, and a new sex cell will be born no more than once every sixteen divisions. While there are few "egoists", they are useful for the colony, its future is in them, one can say – only their overabundance is harmful.) At this stage, group selection will again be replaced by individual selection, because the group will turn into an organism.

In the social groups of primates, the situation is different, but in some ways similar. Paleoanthropologists claim that the situation described by Jack London in the story "Before Adam", when one scoundrel terrorized his entire tribe, could hardly have taken place. In fact, in primitive tribes there was egalitarianism – the enforced equality and restriction of the arbitrariness of the individual fittest within the group. (It is appropriate to recall here that any complex system develops mechanisms that control the behavior of group members. Take, for example, the immune system and apoptosis, which destroy cells that are unnecessary or harmful to the body, or the attitude of worker bees to sisters who shirk work.)

By the way, although the members of the group "on average" will always be genetically closer to each other than to representatives of other groups, it is very unlikely that all members of the group are close relatives. And even if they were, it would hardly benefit them: marriages between relatives lead to degeneration. Most likely, even in a small human tribe, not only siblings, but also stepbrothers, cousins, and fourth cousins, and not relatives at all, went hunting together, and the same applies to women who looked after the hearth and children. However, in order to survive, it was necessary to adhere to the rule: every member of the tribe has the right to count on everyone's help, regardless of the coefficient of kinship and who is stronger. However, a person from a neighboring tribe, under certain conditions, could be considered not just an enemy, but generally a non-human and act accordingly with him...

It can be argued that tribal laws or moral norms are in no way a genetic acquisition, but something from the field of culture. No wonder in the fairy tales of all the peoples of the world it is much more often said that it is necessary to share food and not betray comrades than that it is not good to kill your children or marry your own sister (although there are fairy tales about this too). If unrelated altruism were genetically implanted in us, there would be no need for guidance! But in fact, one does not exclude the other: a predisposition to altruism does not mean rigid determination.

By the way, some experimental data indicate that "altruism genes" do exist in primates. In humans, the innate components of behavior are usually revealed through experiments on twins – so, when solving problems where it was possible to share or not share the points earned with an unfamiliar partner, the attitude towards the partner was much more similar in twins than in non-identical brothers and sisters (so you can't write off this similarity to the influence of the environment). So, the genetic component still needs to be taken into account.

Just us and the marmosets?

If this proof seems insufficient to someone, interesting experimental data on altruism in marmosets Callitrix jacchus have recently been obtained ("Proceedings of the National Academy of Science of the USA", 2007, vol. 104, No. 50). These monkeys do not belong to the higher primates, are not distinguished by high intelligence and are unlikely to tell their cubs tales about greedy and kind brothers. Of course, they probably have training in certain forms of behavior, but in such a situation, which the experimenters came up with for them, marmosets had never happened before. Two monkeys were sitting in adjacent cages, they could not pass anything to each other, but one of them could pull a cart to the cages, the food from which the second always got. And the marmosets agreed to do it. Not necessarily for the sake of relatives – a completely unrelated monkey could be sitting in the next cage. Not because they hoped someday to grab the prey themselves – special measures made it possible to think that the marmosets had fully realized all the treachery of scientists. Not out of boredom – if there was no one in the next cage, they were much less interested in the cart. And not even in the hope of a return favor and not because a neighbor asked them about it! Simply so...

Statistical processing of experimental data confirmed that marmosets have a strange desire to do good to their neighbor, despite the fact that they do not have outstanding mental abilities and they hardly build an internal model of the consciousness of another in their consciousness. (Until now, it was believed that this is what allows people to be truly altruistic beings.) In other primate species (yet?) nothing like this has been found, we with marmosets are unique in our altruism. As the authors of the study suggest, this may be due to the fact that marmosets, like people, take collective care of children, regardless of who is whose: this property is also quite rare. It seems that, speaking of altruism, it is still too early to discount genes and selection.

Of course, when our ancestors embarked on the path of socialization, laws and moral norms had to appear in order to support useful adaptation. However, according to E. Wilson and D. Wilson, the predisposition to cooperation instead of competition, fixed by intergroup selection, was still primary. And in order to serve the new ability for primates to cooperate, all other adaptations that serve communication between people have arisen. And this is our unique brain, and speech abilities, and the human eye, in which the iris and the protein are visible (it is he who gives our gaze a special expressiveness, chimpanzees have completely different eyes), and the ability to point and laugh in company...

It turns out that the influence of group selection on genetics is by no means negligible. And even though the "qualitative leaps" that transform individuals into a group are rare, it is impossible not to reckon with them. "Selfishness defeats altruism within the group. Altruistic groups defeat groups of self-lovers" – this is how E. Wilson and D. Wilson finish their article.

Who is right after all? It's probably too early to draw conclusions yet. Reading Dawkins' rebuke to the Wilsons on the website of the New Scientist magazine, commentators compare what is happening to a boxing match: one fears that science will be knocked out in this fight, the other assures that battles of ideas are always good for science, "and let the best meme win.".. Did the joker remember that, according to the same Dawkins, the "best", that is, the most tenacious, meme is not always the truest?

Portal "Eternal youth" www.vechnayamolodost.ru27.06.2008