Part I: What is evolution?

Common dictionaries provide several definitions for the term ‘evolution‘. Here, we will focus on a very general one, e.g. as taken from the web based dictionary “Webster” ^[42]:

“A process of continuous change from a lower, simpler, or worse to a higher, more complex, or better state.”

To make this more concrete, we need to clarify what is meant by terms like ‘higher’, ‘more complex’ or ‘better’. This can be accomplished by introducing the state variable of entropy.

Entropy:

“Chaos always defeats order because it is better organized.” (Terry Pratchett)

Qualifying entropy

Entropy is a measure for a system’s disorder. For most purposes these two terms can be used interchangeably.

However, unlike disorder, entropy is a mathematically defined concept. Its precise definition is based on the number of states that are accessible to a system. Many possible states mean high entropy and disorder. A derivation of this can be found at other places ^{[43], [44], [45]}.

It follows that low entropy systems can (but do not need to) have sophisticated design and functionality, while high entropy systems generally cannot.

For example a pot of boiling water has rather high entropy, since its molecules hop around to many different places and energy levels. A living human brain, on the other hand, has low entropy, since its molecules are confined to those very few states that create its sophisticated functionality.

It is important to note that disorder is usually more probable than order. This is simply because there are more thinkable states that appear to us like disorder than states that appear ordered, which is described by the second law of Thermodynamics: Every process that actually happens in our universe is accompanied by an increase in the universal entropy.

A good introduction to the second law can be found on the internet ^[43.1] or any local library.

But all over the place we see humans grow and differentiate out of single cells. Is that not a slap in the face of the second law? In fact not. To grow, humans need many low entropy resources to be turned into high entropy ones. Beautiful Linzer Cakes are quite literally turned to crap. So the universal entropy still increases, even if it can decrease locally.

What about the earth? It developed from a lifeless sludge ball into a beautiful environment full of intelligent and charming low entropy systems. What did we do to increase the universal entropy? The fact is that we are powered by solar energy. Without the green plants using this energy (in a process called photosynthesis) to build up our daily supply of low entropy macromolecules, life would not be possible. The sun on the other hand is an enormous entropy fabricant. Its radiation energies blast out into space ready to confuse any little order that might be in their path. The effect is weak but spread on a huge spatial volume. What the second law tells us is that this entropy increasing effect must be greater than the local entropy decrease on earth, or the process would not happen.

It is possible to interpret the domination of the probable over the improbable (i.e. increase of entropy) as the prime motive force for all happenings in the material universe. Energy then is only a device by which entropy transports and delays.

Evolution at work:

“There is nothing permanent but change.” (Heraclitus)

The evolutionary mechanism

While our universe rushes towards ever more entropy, following the second law, few processes have the property to reliably reduce the entropy of certain systems, dissipating it so that the universal entropy can still increase. We call these processes Evolution.

It is intuitively understood that living requires high functionality and thus low entropy. The atoms of a human body could have many chaotic, decaying and dead states, while only very few specific arrangements can create our sophisticated living metabolism. To maintain them over a long period of time, the accumulating entropy must be dissipated in some way. That is why people who like to live very long or forever should have a general interest in the evolutionary mechanism.

The problem with low entropy systems is that they are so extremely improbable. Without evolution, when accidentally something of marvelous functionality and design comes into existence, our universe will expose it to unceasing entropy bombardment. Uncountable numbers of kinetic impacts, light waves, chemical attacks and thermal movement will leave their effect on our system. Because it is environment-dependent and random, some of this effect must impair functionality. A system may self-repair, but any self-repair mechanism can come under attack as well. Given enough time, every non-evolutionary system must decay and perish.

Evolution kicks in when a low entropy system accidentally learns to replicate itself. That is what they are especially good at, because the low entropy can be used to bring about the functionality required for replication. This happens rarely by accident, but it can happen. Random attacks from the outside world work for us this time. Sooner or later they will produce a replicative system, provided that replicators are possible at all. The resources necessary for replication must be drawn out of the environment, ultimately increasing the environmental entropy.

It follows again from random behavior of the universe that errors can occur during each replication. Thus, most daughter systems will have slightly increased entropy, but a few lucky ones will have lower.

Now those few can use their low entropy to become many. With lower entropy allowing for higher functionality, the low-entropy daughters can replicate more effectively than their high-entropy brethren, especially under the pressure of resource limitations. All this leads to increasing numbers of offspring with increasing replicative ability and decreasing entropy.

We come to conclude that evolution creates a local decrease of entropy, which is applied to optimize a replicating system’s power. We get evolution when we have the three Darwinian terms together: Replication, mutation (which should always occur to some degree in our universe) and selection.

Part II: The evolution of life and death.

The beginning

“There is magic in every beginning.” (Johann Wolfgang von Goethe)

The soup

The earth’s virgin oceans apparently have consisted of water under an atmosphere of so called reducing gases like methane, ammonia and dihydrogen.

When triggered by electrostatic discharge (e.g. as provided by lightning), these substances can associate to form larger organic compounds and even biomolecules like amino acids. This has been simulated in the lab. (Miller-Urey experiment) ^[2]:

The earth’s early atmosphere probably did not contain oxygen, which could have prevented the reaction, by quickly reoxidizing intermediate products. Oxygen was first produced by algae and plants doing photosynthesis. Today the reaction pathways by which these accidental-seeming syntheses take place are quite well understood. ^[85]

The idea that similarly to this lab experiment, a solution of various organic compounds formed in the oceans is called the “primordial soup hypothesis” ^[5].

Catalysts

Among organic substances (and thus also in the primordial soup) there are many catalysts. A catalyst is a substance with certain chemical properties that speed up reactions of various other molecules (the ‘substrate’ molecules). The catalyst itself is not consumed in the reaction of the substrates.

For example, platinum catalyzes the burning of hydrogen in air, acid catalyzes the decomposition of starch into glucose, and a molecule called ‘sodium-glucose-symporter’ catalyzes the transport of glucose through intestinal cell membranes. Without the catalyst, all these processes would be far too slow to be seen happening at all.

With a random cocktail of catalysts in the primordial soup overall reaction speeds increased and even more variety emerged.

Autocatalysis and Replicators

An interesting example of a catalyst is the manganese ion. It catalyzes the reduction of permanganate that yields more manganese ions. As long as permanganate resources were abundant, it could be said that manganese ions catalyzed their own production. The reaction is called autocatalytic. In fact, an exponential increase in manganese ion concentration over time is observed.

To call a molecule a true replicator it could be required that it does not just cleave a more complex structure into more of its kind (as is the case with the reduction of permanganate), but that it catalyzed the assembly of copies of its copies out of simpler building blocks.

It is imagined that in the primordial soup exactly such replicators must have formed by accident.

Of course, their numbers started growing at once.

The arms race

Soon replicator molecules were all over the place, using up organic resources faster than they were recovered. Different sorts of replicator molecules started to compete for resources.

Clever designs developed the ability to synthesize building blocks literally out of thin air and sunlight in a process called photosynthesis.

Others may have found ways to chemically disassemble competing replicators and assimilate the released building blocks.

This in turn created a selective pressure: Replicators that were able to defend themselves against disassembly were at an advantage now. (They had not been before the advent of disassemblers.)

An arms race may have resulted that drove towards ever more sophisticated chemical weaponry and armor.

Whatever the details were, it is clear that only the most effective designs could persist (selection) and ever more effective ones emerged through beneficial errors in the replication process (mutation). The conditions had met for evolution to start off.

RNA

When one wonders what may be the chemical requirements for an autocatalytic replicator molecule, it comes down to the following: The molecule must both be good at storing the information about itself that is to be replicated, and good at executing the creation of its copy, that is to be a good catalyst.

This conflict of interests is exemplified by three specific contradictions ^[11]:

A good information storage is a molecule with an easy to read linear makeup, while a good catalyst is better served with a complex three dimensional structure to achieve the necessary functionality ^[12].

An information storage should use few distinct building blocks to represent code-like information and allow for error-correction procedures ^{[13], [14]}, while a good catalyst should use many building blocks or arbitrary design features to achieve the chemical variety that is necessary for catalysis.

When suffering mutations, an information storage must retain its physical properties, so that it can still be read and replicated by existing mechanisms, while a catalyst suffering mutation should change its physical properties so that it can evolve new or better functionality ^[11].

One molecule has long been supposed to be a sort of compromise between these conflicting demands. It is known as ribonucleic acid, or RNA ^{[15], [16]}.

The compromise is realized as follows:

RNA is a linear molecule that can, however, fold in complex three dimensional ways. Its information is encoded in the sequence of four distinct organic bases. The folding does somewhat depend on the base sequence, but allows for a good deal of diversity before it collapses ^{[10], [12]}.

RNA was shown capable to perform catalytic action ^[6]. But its full potential became clear only recently, with the discovery of ribozymes ^{[7], [8]}. Ribozymes are powerful RNA based catalysts that can even work on RNA substrates, which brings them already close to autocatalysts.

To get a notion of how powerful a ribozyme can be, let this one picture be worth a thousand words:

^{[75], [4]} This is an X-Ray diffraction picture of a so called hammerhead ribozyme (strings and wire frame) working on several substrate molecules (colored spheres), to bring them together in positions and angles favorable for quick reaction.

Ribozymes were found in several viruses, algae and plants, as well as evolved artificially ^{[8], [9], [10]}.

The building blocks of RNA have not all been synthesized in a Miller-Urey like experiment, but given the building blocks, their assembly has been achieved using clay catalysts ^[3]. Thus the postulate of early RNA based replicators is not fully sound. But given the above considerations, today it is widely agreed among molecular biologists that the history of life as we know it was preceded by an ‘RNA world‘, where both information storage and catalytic action were based on RNA ^[17].

Further support comes from the fact that today’s organisms still contain much chemical equipment that resembles RNA, especially in evolutionarily old and unchanged components ^{[18], [19]}.

Division of labor

Despite all its success, RNA was still a compromise to achieve genetic and catalytic goals at once, which are complementary in principle. RNA can achieve both goals reasonably well, but none in perfection. It would be more efficient to evolve one molecule specialized in genetics working hand in hand with a different one specialized in catalysis.

Of course, this is precisely what happened. A transition was made from one-biomolecule replicators to two-biomolecule replicators.

RNA may have been a necessary step to evolve the complexity required for the machinery that can keep two biomolecules together (e.g. in a cell) and that can translate them into each other. But once the powerful two-biomolecule system was there, it promptly consumed all RNA competition.

The two biomolecule system is still dominant today and will be characterized in the following.

Genes and Proteins:

“Rain falls, wind blows, plants bloom, leaves mature and are blown away. These phenomena are all interrelated with causes and conditions, are brought about by them and disappear as the causes and conditions change.” (Buddha)

Structure

The part of the information storage is played by deoxyribonucleic acid or DNA. This close relative to RNA still codes information in the sequence of four distinct organic bases. A minor modification was made that interdicts arbitrary folding, but coils up two strands of DNA in a regular double helical form that ensures smooth readability.

The part of the catalyst is played by protein. Proteins are linear molecules, too, but made up of amino acids (as we know from Miller-Urey ^[2], these were probably an abundant resource in the earth’s early oceans). The 20 distinct amino acids used in proteins have a great chemical variety that grant the protein diverse catalytic function, along with the ability to bind foreign catalysts (e.g. metal ions), to gain further functionality.

Also due to the variety of amino acid properties, the protein molecule can curl up into any thinkable three dimensional conformations. Even minor changes in amino acid sequence can hereby lead to a very different conformation in space and thus very different physical and catalytic properties, which ensures good evolvability.

The protein catalysts in use by today’s organisms are called enzymes. They are generally bigger and much better organized than ribozymes. (Note how the term ‘ribozyme‘ was synthesized from the previously known ‘ribonucleic acid‘ and ‘enzyme‘.) To get an impression of what today’s state of the art protein enzymes look like, see this beautiful x-ray diffraction picture:

^[75]

This shows human hemoglobin A, responsible for carrying oxygen from the lungs to the tissues with the blood stream and carbon dioxide back. Oxygen is bound by the four yellow spheres (iron ions) inside the green bubble fields (heme groups). Carbon dioxide is bound by four of the eight the string ends (can you find them?). Also serves as proton buffer along the way. Individual amino acids are not shown. Instead their 3D-conformation is designated by the strings and helices. A total of 574 amino acids on four intertwined protein strings make up the molecule. There are several thousand trillion (10²¹) hemoglobin molecules in an adult human’s bloodstream.

Genes: The relationship of DNA and protein

Proteins are made from the base sequence information in the DNA. (Interestingly enough, the building of proteins according to DNA information is accomplished via RNA intermediates.) The protein in turn can catalyze the replication of the DNA or any other process that might be beneficial to it.

As there are only four distinct bases in the DNA, but 20 distinct amino acids, it is a triplet of DNA bases that is translated into one specific amino acid. Thus the DNA has thrice as many building blocks as the protein it codes for.

A piece of DNA that codes for one protein or several related proteins is called a gene. Of course, evolution drives towards genes that code for proteins good at replicating the genes. (Such genes will be abundant, because they get replicated. Others not.)

The following is an attempt to interpret all life on earth as increasingly complex efforts of diverse genes to make proteins good at replicating them ^[24.1].

Genetic coevolution

We can now imagine how genes are selected by the inanimate environment. Another aspect of the environment has been ignored so far. This rather crucial aspect is other genes. It shall be addressed in the following.

It is important to know that there can be arbitrarily many genes on a single DNA molecule. Thus one gene can directly profit from the protein created by a neighboring gene. This gives genes the possibility to ally, much unlike their autocatalytic RNA predecessors.

How can such an alliance come about?

Imagine a gene that is very successful, a very good replicator. Let us call it gene A (A for ‘attractive‘). Gene A fabricates some protein that is extraordinarily good at replicating DNA under the circumstances given.

Another gene, for example gene F (F for ‘free rider‘) is on its own less capable. Yet if gene F made a protein that integrated F into the very same DNA strand that A is on, it would get replicated along with A, by A’s powerful protein machinery. We should expect genes to evolve that identify and attach themselves to successful genes.

Of course, with an entire primordial ocean of genes inclined to integrate themselves into A’s DNA, A’s success will soon crumble. For A’s DNA strand will get longer and longer. It will need more and more resources to replicate all the free riders, which makes A inefficient.

So mutants of A that have mechanisms to defend against free riders should be at an advantage now. For example A could make a protein that degrades all DNA that does not display certain special structural features that A has. An arms race may result that makes both A and F ever more complex and effective.

Let us now consider a different encounter. Gene B is successful for some other reason than A. Let us say A’s protein is successful at the acquisition of some rare organic resource, while B’s is successful at blocking off all free rider DNA. Of course, B is as inclined to get on A’s DNA as is any gene because of A’s good protein. Mutants of B that block off all free riders but make an exception with A will replicate more effectively because they from now on fabricate A’s resource acquisition protein.

But A, on the other hand, is equally inclined to get B on its DNA. Those mutants of A that let B attach to their DNA will replicate more effectively, because they will from now on fabricate B’s protein that blocks off free riders.

No matter if we choose A’s point of view or B’s, the result is a combination of genes AB that unites the beneficial properties of both. Such a mutually adapted gene complex is called in short a ‘geneplex’.

Geneplexes can get really huge, comprising ten thousands of genes that give rise to such complicated organisms as humans. But more on that later.

For now it will be enough to summarize that all genes tend to attach themselves to successful (attractive) other genes, while attractive genes evolve to keep less attractive ones out.

This is observable whenever genes are exchanged between DNA molecules. E.g. cells of all sorts actively defend against viral infection and most sexually attractive humans thoroughly avoid mating with significantly less attractive ones ^[20].

Sex

All higher forms of live have evolved some mechanism to exchange genetic information between individuals, also known as ‘sex‘. This is mainly due to the great possibilities that this opens for the elimination of disadvantageous mutations and DNA repair. When the DNA of one organism has become compromised, it seems sensible to use DNA from another organism of the same species as backup copy ^[21].

On the long run sex means also easier adaptation to changing environments: Genes that induce sexuality will find that their surrounding geneplexes will be quicker at catching up new advantageous mutations that happened only in single organisms. So geneplexes that readily ally with sex inducing genes will be at an advantage, which also propagates the sex-inducing genes themselves. Sex spreads ^[21].

Sex is observable in its most original form in bacteria. They exchange so called ‘plasmids‘, small DNA molecules with a few genes on them. For example, when a bacterium somehow acquires genes for resistance to some antibiotics, it can package them onto a plasmid and send them to fellow bacteria that gain this ability on receipt ^{[29, 30]}.

The sexuality-inducing gene protects its own copies in other individuals from antibiotics, respectively accesses the protection present in other individuals, and should therefore proliferate.

Sexual reproduction

In many organisms, sex is coupled with reproduction. That is, the DNA from two mating parent organisms joins to form a random combination on that the offspring organism is based.

Thereby, each DNA error that may have been present in one parent is possibly compensated by DNA from the other parent. In the same way, beneficial mutations from one parent have a chance to be present in the combined DNA.

Of course, with a random combination of genetic information there will also be offspring that still contain the error or the unimproved version of the mutated gene. Yet those will be out-competed by offspring that do not.

Thus we can conclude that offspring of sexually reproducing organisms are on the average improved and error-corrected versions of their parents.

Mate selection provides an additional mechanism to improve the offspring. When individuals have evolved high enough to screen their mates and judge them for genetic quality, they can choose their mate so that error correction and acquisition of beneficial mutations will be maximum. More on this will be said later.

An exercise in vocabulary

By now, you should have been able to spot the way thought is made in evolutionary genetics. In principle all evolutionary happenings could be discussed like this. But when talking about more complicated matters, being faithful to the correct evolutionary way can soon get very lengthy and exhausting. To my knowledge it was first observed by the evolutionary biologist Richard Dawkins ^[24.2] that an informal language on a higher level can be used as a deliberate shortcut.

In high level language we could for example claim:

“The genes wanted humans to walk upright.”

Of course, that is simply wrong. Genes are structural sequences in the DNA molecule and as such incapable to exhibit anything that could be called a ‘will’ in analogy to human will.

Yet the statement can easily be back-translated into correct language:

“Geneplexes that produced a tendency in their carriers to walk more upright found themselves at a selective advantage as more of their carriers reached sexual maturity because they had a better stand in fights and could reach for food higher than their competitors. That is how such genes came to proliferate.”

You see that the high-level statement really makes things a lot more convenient to talk about. When using high-level language, we must always keep in mind that our argument remains back-translatable or we can get completely feet-off-ground.

Some other valid high level expressions are used in the following statements (it may be a good exercise to back-translate them on your own):

“Fish developed hydrodynamically shaped fin in order to swim as fast as possible.”

“Life left the oceans to claim the uninhabited lands.”

“Survival of the fittest.”

The back-translation of the latter deserves a closer look. This expression is attributed to Charles Darwin ^[1] and is still considered to summarize the fundamental mechanism of evolution. When we back-translate it, it comes down to something like “Genes that fit the demands of survival, will survive.”, which is not at all very informative.

More recently, Darwin’s statement has been reformulated as “Genes are selfish.” ^[24], in the sense of “Genes that have nothing but their own replicative power thoroughly optimized will be the most abundant.”, which seems also quite needless to say.

In this sense the fundamental evolutionary mechanism is not a natural law, nor an evolutionist presumption, but a tautology. Its truth is unconditional on any information and logically compelling.

It may be educative to consider a high level language statement that fails to match realities, because back-translatability is lost:

“Evolution made humans capable to invent high tech weaponry to ensure our dominance on the planet.“

In fact inclination to military research appears to be a by-product of genetically advantageous brain developments that might have been advantageous as long as swords and spear-throwers were concerned. (More on that will be said in the memetics chapter.) Modern weaponry, on the other hand, could easily have meant the end of all evolution on the planet. Nuclear weapons may serve as an impressive example that evolution can not look into the future to assess the desirability of its creations. Unlike humans, evolution knows nothing about desirability. What happens happens. More is not said and shortcut language must be used with caution.

Now we have the equipment to advance on more specific issues. In the following, shortcut language is used when appropriate without further indication. Occasionally, when of special concern, an explicit back-translation is made.

Higher organization:

Cells

As was already suggested, some genes came up that made proteins that surrounded the DNA with the phospholipid shield called a ‘cell‘. Cells are in fact marvelous survival machines that conveyed a deluge of advantages to the genes inside.

They provided a constant and optimal working environment for proteins, shielded the DNA from damage, gave convenient shape for motion and allowed the storage of resources and various maintenance molecules. Inside cells ever more complex and powerful chemical facilities could be installed. 

Genes that readily organized themselves in cells outwit almost all competition. (We still have the non-cellular DNA of viruses today. However, they draw on the machinery of a host cell for their replication.)

Multicellulars

Some geneplexes initiated the creation of entire non-replicative cells (the so called soma cells) only for protein production and maintenance of the replicative cell (the so called germ cell). They found themselves in a favorable position and thus multicellular organisms emerged.

The soma is created in a particularly elegant way. The germ cell simply does what it is especially good at: it replicates. In a sophisticated process called mitosis or cell fission, new soma cells are formed out of the germ cell. (In macroscopic organisms the exorbitantly high numbers of soma cells are achieved via an intermediate tissue, the so called stem cells. They are replicative themselves, giving rise to an exponential growth of soma cells. Once the individual is mature, stem cells are diminished and only used for special purposes like wound healing and redundant germ cell production, but that is not the issue here.)

What is important is that due to their mode of creation all soma cells contain the same genetic information as the germ cell that spawned them. They vary only in instructions on what parts of the DNA information actually to execute (that is build proteins from it). Yet the total genetic information is the same in all soma cells and the germ cell.

Death:

“Being born is not a crime, so why must it carry the sentence of death?” (Robert Ettinger)

What is death?

All multicellulars are mortal. That is the soma cells of any one individual die after a finite and more or less fixed time span. Most die in a complex and lengthy process called aging, which gradually deprives the organism of its functionality. It is rendered increasingly vulnerable to attacks from the environment, so that it must perish sooner or later. The genes emerge in just a few youthful germ cells that form new soma and new organisms to aid them in their next replication process.

How does this come about? If the germ cells can be maintained in a youthful state, why can’t the soma cells? Would it not be advantageous to genes if they made ageless soma that would have indefinite time to replicate the genes over and over again? Furthermore, the mere maintenance of the soma should be both cheaper in terms of metabolic energy and mechanistically easier than its periodic reconstitution.

But in fact there the vast majority of multicellular genes refrained from immortal soma, so there must be a hook to it.

The genetic advantage of somatic death

When multicellulars engage in sexual reproduction, only two germ cells do the entire job. They fuse their genetic information into a combination cell that is, as was argued above, improved in fitness.

At the same instant the old soma is obsolete. For it still contains the old information! Sex would be pointless if there was no way to get the new information into each and every soma cell. For they are supposed to create upgraded protein machinery using the newly acquired genetic information. If the old soma was allowed to persist, there would be no use the acquisition of new information in the germ cell at all.

The most successful way to get the new information into the soma was to dispose of all the old soma cells and to recreate them from the scratch through a new series of cell fissions. That is the death of the organism and the growth of a new one.

Frequently old soma is used to provide the necessary resources. E.g. in mammals much of the soma production takes place right within the parent individual, drawing on her body resources. Even later the old soma functions to support the offspring. No need to tell a human that parents do all they can to get their offspring ahead, while they themselves are slowly shut down. It is now understandable that genes that induce this behavior simply fare best.

Genes that induce aging and death will find themselves more frequently in error corrected and better adapted offspring organisms, while genes for ageless soma would tend to get stuck in theoretically long-lived but practically maladapted parents.

It was to combine the great advantages of both multicellularity and sexuality that made individual death a survival trait for genes ^[22]. Today the vast majority of all multicellulars do it this way.

Support from Biology

Unlike the popular opinion, we are lead to the conclusion that aging and death are not inevitable results of disorder and random deterioration but programmed functions of sexually reproducing multicellulars. Although it is rather ancient ^[50], this hypothesis has seemingly never attracted much public attention.

Immortal soma would unquestionably be advantageous to genes. But it is even more advantageous to upgrade and improve the soma from time to time, discarding the old.

Support comes from empiric facts, such as that most unicellular organisms, all asexually reproducing multicellulars and germ cells are ageless.

An especially well suited example are many species of the bamboo. They reproduce mainly asexually via underground rhizomes and thus the individual plant is ageless. Yet very rarely sexual reproduction through flowering takes place, with frequency varying with species from three to 120 years. All plants of the species all over the globe flower simultaneously – how coordination is achieved remains a mystery.

Immediately after the emission of huge amounts of seeds, all adult plants undergo rapid death and decay to make room for their offspring with its ‘better’ recombined genetic information. In fact this event is something like a catastrophe among hobby bamboo cultivators ^[23].

Another overdue example is the female pacific salmon. Salmons live and develop in the oceans and enter the rivers only for spawning. Salmons know no child care, thus the parent is superfluous once the eggs are out. Just after spawning, a metabolic shutdown program is activated that cripples and kills the fish within a few days. During this process salmons show all major symptoms of vertebrate aging ^[34].

Young salmon have even been shown to obtain resources from their parents’ corpses ^[35], which is possibly a key genetic advantage of instant parental suicide.

And what about humans?

Following the amazing mechanisms of sexual recombination, evolution produced geneplexes of ever higher sophistication. Today, huge populations of allied genes move about in the trillion cell heavy somatic robots that call themselves humans. Our soma is highly differentiated into epithelia, tissues and organs to bring about our marvelous functionality.

Finally, as a center of input processing and motion control, our brains evolved. They allowed us to scan the world from the tiniest quantum particles up to the rim of the universe. Yet an object of even remotely comparable complexity and marvel we have not found.

In those brains something very odd and possibly unprecedented happened. Through some unknown miracle our minds came about. Not foreseen by the genes, their robots learned to identify themselves as persons, as ends in themselves.

But still the most efficient mechanism to make use of sexuality is to decompose us along with each information exchange.

Widely out of genetic control as we find ourselves today, we need not agree. Today we are persons, not mechanisms. Who should forbid us to see more than an eye blink of the universal theater that created us?

Evolution equipped us with the power to take our lives in our own hands today. Maybe we really have a chance. It would be a shame not to have tried.

On the feasibility of life extension

When we understand aging and death as programmed functions rather than as inevitable by products of life, we should expect to find distinct mechanisms that realize this program.

In fact a likely candidate for such a mechanism has recently been discovered. This mechanism is called telomere shortening ^[51]. Only a very short summary is given here. For full information see other sources ^{[55], [57]}.

Telomeres are special regions at the end of our DNA molecules that for certain reasons shorten with each cell division, like the string of an infant’s musical clock. Once they are critically short, the cell stops normal function and enters senescence.

With average telomere length being strongly correlated to cell donor age ^[52], it seems plausible that increasing amounts of senescent cells account for the senescence of an entire human.

Germ cells, however, can circumvent the telomere shortening mechanism. During their development, germ cells must rejuvenate or the embryo would be biologically as old as the humans who conceived it ^[51].

Rejuvenation is accomplished by an enzyme called telomerase. It works by simply lengthening telomeres to their normal size ^{[46], [53], [54]}.

When telomerase is applied to senescent human soma cells in a Petri dish, they quickly regain youthful telomere length and vigorous function.

As simple and straightforward as all this may seem, scientists are far from able to rejuvenate entire humans with telomerase technology. There are plenty of complications associated to telomerase application in living organisms, and probably other aging mechanisms that have little to do with telomeres ^[56].

From the existence and our partial understanding of distinct aging mechanisms we can only derive the hope that aging may in fact not be as inevitable as in our myths and that the work necessary to achieve a vast extension of the human life span may be finite.

Part III: The evolution of society and personality.

Memetic evolution:

“A scholar is just a library’s way of making another library.” (Daniel Dennett.)

In 1989 the evolutionary biologist Richard Dawkins pointed at another system in that evolution may occur, quite beyond pure biology ^[24.3]. That system is human thought, habits and culture.

What is the underlying replicator unit in that system? When a human communicates a thought and another human catches it up correctly it can be said that the thought replicated from brain to brain. A piece of thought that can replicate this way is called a meme. (In phonetic analogy to gene.)

Dawkins’ somewhat vague base idea is elaborated by Susan Blackmore in her 1999 book ‘The Meme Machine’ ^[25]. Much of the following chapter is actually based on this book.

Of course there are other models of human culture that may be equally valid. Yet I feel that the evolutionary approach has an especially charming appeal and in some ways unique predictive power.

Mainly because of its relevance to the life extensionist’s social interests (more on that later), the idea will be outlined here. If you are interested in more detail I recommend Blackmore ^[25] or other literature on memetics ^{[47], [48], [49]}.

Replication

When biological evolution had led to the first humans capable to invent tools and primitive technology, a new style of learning became important. It is inefficient to invent everything from the scratch when many beneficial technologies are already available in other humans. So genes that favored the traditional conditioning procedures were out competed by those that additionally granted their humans the ability to overtake technologies by imitation.

The spread of the ability to learn by imitation among early humans provided the ground for the new replicative unit that is called the meme. It is a unit of humans thought, behavior, technology and culture. A meme is just any piece of information that is transmissible through imitation. Memes are for example the instructions how to build and apply the wheel, the Aborigines’ way of body painting, the latest style of clothing among American youngsters, a song by your favorite pop star or the headline of today’s newspaper. Memes replicate from brain to brain during every conversation, public speech, newspaper read, television broadcast and internet browse.

Genes can only be transmitted from parents to offspring (Bacteria provide an exception. With plasmids, they can exchange genes between individuals independently from reproduction and even between different bacterial species). Memes on the other hand can spread from individual to individual without regard to familiar bonds. Recently, with mass media becoming available, memes discovered ways to copy into most humans on the planet instantly. To get into the media is one of the greatest selective advantages that can happen to a meme.

With mass media being a product of memetic evolution, one can say that memetic evolution accelerates itself in a way much faster than its genetic counterpart.

From emergence of the ability to learn by imitation to today’s technology and culture only a few million years passed. For a similar increase in complexity billions of years had to pass in the early stages of genetic evolution.

Mutation

Just like with genes, mutations occur in the replication process that alter the meme’s replicative success. For example if somebody gets it wrong during the transmission of the instructions to build a longbow, most probably instructions for something useless or even dangerous will result. Though it can happen by accident or planning of the transmitting humans that the result is an improved longbow or crossbows or whatever.

The rate of memetic mutation appears to be high in comparison to genetic mutation. For example during the telling of a story with every replication several words should be altered due to human creativity and imperfection of memory. DNA on the other hand is replicated with a fidelity of less than one error in a thousand billion (10¹²) bases.

What is more, the mutation of memes is partly directed and sped up by human planning, while genetic mutation is still mostly random.

Selection

The selective environment comes from humans’ limited meme storage and transmission capacity. Many more memes are coined than can possibly be spread and kept in humans‘ brains. They have to compete for our storage and transmission resources. Through evolution they acquire ever more complex mechanisms to attract our attention. The following two paragraphs will cover two important selective forces.

Gene-meme coevolution

Genes and memes evolve in parallel, one being part of the other’s selective environment.

Especially the genes for characteristics of the human brain are important factors in the selection of memes. During the embryonic development of the brain structure, genes have plenty of possibilities to affect the way we catch up memes.

Some memes may be especially helpful for human survival and reproduction. Genes that produce an inclination towards such memes should be at a direct selective advantage. This has been described in the past, frequently in non-memetic terms ^[31]. For example it is often said that male youngsters have a special affinity for weapon and sex memes, while their female contemporaries tend to prefer clothing and beauty ones.

In general, it is intuitively understood that a meme ‘of use‘ to humans is a more successful replicator than its nonsense-containing kin. The instructions for the most accurate and far reaching longbow is likely to out compete all those instructions for dangerous wooden things that crack and break in your hand. Genetic evolution equipped us with the (at times somewhat faulty) ability to identify memes with a certain ‘usefulness‘ pattern and favor them in our selection.

Meme-meme coevolution

As with genes, an important factor in the selective environment of a meme is other memes.

Imagine for example the two memes “Carry dry wood to your cave.“ and “Strike a piece of metal with glittering Stones.“. None of them makes sense on its own, but together they allow humans to create campfires and thus they have an enormous selective advantage. Similar to genes, memes can ally to form coadapted meme complexes (memeplexes).

As their biological kin, memes coevolved into the most complex and amazing structures, such as religions, economic systems or the theory of quantum physics.

It is almost no surprise that these sophisticated cultural ‘organisms‘ contain traits highly analogous to the replicative apparatus, metabolism and immune system of a living cell.

Imagine the European medieval Catholic Church. It entertained the inquisition, a merciless kind of immune system. Once its attention was drawn to heretic memes it began to eliminate their carriers without hesitation. Similar systems can be found in certain absolutistic countries that pursue carriers of every meme hostile to the national ideology memeplex.

A very different kind of immune system is part of the recently evolved memeplex of the sciences. It makes its carriers strongly oppose any meme that is foreign to science because it is untestable or can be proved wrong. It does not need to shed any blood of disbelievers, but rewards believers with high usefulness.

The power of Memes:

One comes wonder how memes can have such a great impact on our lives and even lead us to actions that seem disastrous to our genes (e.g. martyrism, celibacy or birth control). Would not genes that allowed such behavior be outwit by genes that forbid it?

Once we accept that memes undergo evolution, it does not seem irrational to expect that they might evolve enough functionality to simply exploit human brains for their replication, overthrowing any attempts of our genes to prevent this. Because of this, memeplexes have been compared to biological viruses ^[86]. But what could be their actual mechanism to do so? A possible mechanism is provided by a model called runaway sexual selection ^{[26], [27]}.

Sexual selection

This model will be introduced with an example from biology: The peacock’s tail. Already Darwin found that such highly exaggerated biological structures exist, that must be detrimental to individual survival. He concluded that features that enhance an individual’s sexual attractivity will give it more offspring and be selected for, even if they endanger its survival. ^[33]

In Peacocks, the male bird has to spend much metabolic energy to build up its tail feathers’ sophisticated ornaments. Moreover, the huge colored feathers must help predators to spot the bird and slow it down when it tries to escape. Both effects undoubtedly reduce genetic fitness, but peacock tails do exist. So how are these disadvantages compensated for?

It is imagined that in some past day, a moderate increase in tail size must have been advantageous. For example to stabilize flight through windy air this may have become necessary. Consequentially big tails became sexually attractive. Females that preferred the male with the biggest tail they could find got fitter offspring and the genes that induced this preference proliferated.

Thus males could increase their tail-size even beyond what was good for individual survival. Huge-feathered peacocks may often fall prey, but still their genes spread effectively because they attract many peahens and have plenty of offspring.

This entails another question with a less straightforward answer: Why are peahens still attracted by exaggerated tails? Would not genes that attract peahens to decently tailed cocks now spread better, making the exaggeration obsolete?

No, simply because there are presently many other females around that find big tails attractive. When a single mutant peahen is attracted by decent tails rather than exaggerated ones, her sons will inherit decent tails from their decently tailed fathers. But that makes them unattractive to most contemporary peahens! Thus they are unable to spread the genes that caused mother peahen’s shift in attitude.

Instead, genes that attract peahens to huge tails will give them sexy sons and thus proliferate.

The process reinforces itself, creating exorbitant costs for the male feathers. Decent tails have become unsustainable. The species is trapped in a vicious circle.

Imitation as our peacock feather

Now familiar with runaway sexual selection we can apply it to the human ability to imitate and solve the initial question about the power of memes ^{[28], [25.1]}.

As was already suggested, during the development of Homo sapiens the manifold of technological inventions made the ability to learn by imitation a survival trait.

Thus this trait also became sexually attractive. Genes that produced an inclination to mate with good imitators were at a selective advantage because they tended to end up in well-imitating offspring ^[25.2].

Once many humans found imitators attractive, genes that made their carriers show off their given ability to imitate were at a survival advantage as well, because they were found more attractive and got more mates. Even exaggerated, wasteful and dangerous displays of imitative ability became tolerable, as to make no mistake in attracting mates compensates for a lot of individual disadvantages. The species got trapped in the same vicious circle as peacocks and peahens. Being attracted to excessive imitators makes attractive offspring, anything else is unsustainable and imitative ability is boosted to astronomic dimensions.

Genes for strong imitation have so huge self reinforcing advantages that some of their carriers may even imitate behavior that makes them self-sacrifice or cease to reproduce. On the average, the attractivity of the genes’ other carriers will compensate for such losses.

Of course, stone-age people most probably didn’t make all these complicated thoughts before choosing a mate. It’s just that genes that directed their carrier’s feelings towards people that showed off extreme imitative ability simply were at a sexy-offspring advantage. The mechanism works with nobody thinking about it more than a woman peacock when she is attracted by her mate’s tail.

Experimental support

This speculative theory is supported mainly by the far reaching analogies between the peacock tail and human imitation, suggesting a similar origin. The human ability and eagerness to imitate, just like the peacock’s tail, is a highly complex, exaggerated structure that appears to carry direct and heavy genetic disadvantages. As the peacock’s detriments are obvious, we will now investigate the human ones.

Brains are difficult to build, especially since neural conductors require a fatty isolation layer called myelin. Fat on the other hand is our most effective energy supply and thus very expensive when used as construction material.

Also because of the risks it carries at birth, there has been much wondering about the big brain.

The main problem, however, is the great metabolic cost that brains carry ^[36].

It is often said that the human brain consumes 20% of our basal energy demand, while in our close relatives the chimpanzees it is only 8% ^{[37], [25.3]}.

The one major difference in performance between human brains and other primate brains is that we alone are capable to learn by imitation to any significant degree. Primates can use tools and various fancy learning techniques for their application, but they are very weak at imitation ^[38].

Then we should expect that imitation uses those parts of the brain that are especially enlarged in comparison to other primates, which are the frontal and prefrontal cortex ^[25.4].

Recently developed PET scanning techniques were able to confirm this to a substantial degree ^[39].

When we compare the magnitude of other primates’ imitative ability to our own, this may be a similar ratio as when we compare the average bird’s tail feather to the peacock’s tail.

If imitation really needs so much energy and computer power, why does it? Probably because the underlying mathematics are really tricky. There is not much experimental data available, but so far research suggests that imitation is achieved in distilling a ‘goal’ from the observed action and then derive means for its achievement ^[39]. A complicated transformation must be made from the observed means to the imitator’s own muscle movements that will further the desired goal. A new kind of nerve cell has been discovered that appears to be specialized in precisely this task ^{[40], [39]}. Interestingly, the existence of such a nerve cell had been predicted by memetic theorist Susan Blackmore. ^[25.7]

Tail feathers today

The determination of what is genetically advantageous was given a surprising twist by sexual selection. Only this twist boosted humans’ imitative ability and made memetic evolution such a powerful force in the forge of human culture. If this theory is applicable, then least some remains of imitation’s attractivity should still be detectable in modern societies.

The best imitators should then be very attractive people. But how can our mating drives identify the best imitators? It should be those people that are most frequently involved in the spread and receipt of large amounts of memes. (Especially of useless memes that express the runaway character of sexual selection).

In modern times, this is certainly people with access to the mass media, including movie stars, pop musicians, fashion designers, artists and writers. In fact, persons who are successful at these professions usually have legendary sexual lives. Even unimpressive physical constitution does not seem capable to override this effect, a standard example being Charles Chaplin ^{[41], [25.5]}.

Imitation’s attractivity is also detectable on smaller scales in the followers of fashion. Often when we desire to imitate the latest clothing or body decoration habit, we are well conscious that it is because we hope to increase our sexual attractivity.

However, especially in western civilizations sexual selection pressure has become much weaker. Since around the medieval, the social system allows almost anyone to have a roughly equal amount of offspring without much regard to her or his attractivity. So a growing amount of deviations from the rule can find their survival niches. ^{[80, 81]}

(In my opinion this is one of the few indisputable advantages that civilization brought to the individual human: On the long run it loosened some of our genetic ties. But I am philosophizing right into the blue and that is not the point of this article...)

The ultimate memeplex

The most complex memetic structures that we have encountered so far reside dislocated in our culture, between individuals. Probably no single individual has all aspects of the Jewish religion inside her or his mind, nor all details of quantum physics. Much of the information of such huge memeplexes is externalized from human memory and written in books or stored in computers ^[32].

However, there is one very sophisticated memeplex that resides isolated inside single human minds. This is the sum and interplay of all memes that are part of someone’s personality. It has been called the ‘ultimate memeplex’ or ‘selfplex’ ^[25.6].

Being integrated into a selfplex is greatly beneficial to a meme’s replicative success. For people talk about their own personal beliefs much more frequently and persuasively than about other information. It could be said that it is an attribute of the selfplex to make them do.

Just knowing what vegetarianism is will not make me propagate it a lot. However, if I am a vegetarian myself, I will eagerly defend my views before non-vegetarians and chances are that I will convince some of them to become vegetarians themselves.

It follows from the selective advantage that memes gain inside a selfplex that they evolve to be compatible with many selfplexes. In other words, selfplexes are memetically attractive. Thus we should expect selfplexes in turn to have evolved mechanisms that keep unattractive and incompatible memes out.

Do such mechanisms exist in humans? But of course. When memes assemble into a selfplex, a delicate selection and adaptation process takes place.

Consider the meme containing instructions to slaughter and shear a rabbit. An eremite forest-dweller who comes across this meme will most gratefully pick it up and propagate it among his fellows. So will eventually a city dwelling pubescent boy to impress his friends at school.

On the other hand imagine a female vegetarian in her twenties. Her selfplex will most violently reject the same meme.

Selfplex activities

It is important to note how selfplexes actively defend themselves. Whenever we feel that somebody’s views are clearly at fault and that we must convince her or him that our own attitude is better, it seems to be a valid interpretation that the originator of the action is our selfplex applying the amazingly complex abilities that it acquired by evolution.

Just like the geneplex of a living cell rejects foreign DNA, the selfplex is capable to launch a memetic immune response against incoming foreign memes, at times by driving his human into stubborn and offensive argument.

Another interesting case may be a passionate stamp collector whose selfplex will make him browse the web preferably for information on rare stamps rather than e.g. for astrology.

Stamp information that ultimately enhances the selfplex is welcomed and actively searched, while incompatible information receives no special attention.

When deciding to follow the memetic train of thought, we can’t help but say that our selfplexes are actively working on their own maintenance, extension and the replication of their constituent memes into other humans. In doing so, they exercise substantial power on our behavior.

It might be educative to look at these happenings from a physical point of view. Nobody has yet seen what a meme looks like, but we can imagine