In the jacket of the book The Blind Watchmaker by Richard Dawkins, we can read the following:
“Natural selection, the unconscious, automatic, blind yet essentially nonrandom processes that Darwin discovered, and that we now understand to be the explanation for the existence and form of all life, has no purpose in mind. It has no mind and no mind’s eye. It does not plan for the future. It has no vision, no foresight, no sight at all. If it can be said to play the role of watchmaker in nature, it is the blind watchmaker.”
Darwinism, or natural selection, is the theory whereby if we accept the possibility that small changes are constantly taking place in the genetic makeup of all living organisms (we need not concern ourselves at this point on how those changes would take place), then as those changes begin to accumulate in each individual organism the competition for survival will favor those individuals whose variations have given them an advantage over other members of their same species. Thus, by fierce competition or “survival of the fittest”, Nature selects those individuals and species better equipped to survive in the long run. This in turn favors the predominance of more complex organisms and the extinction of the less fit. So goes the argument of “traditional” evolution.
There is such an overwhelming amount of evidence in favor of natural selection that even among the Catholic Church, a long-time bastion of creationism, eminent theologians are now open to discuss and accept the possibility that if a supreme being created all of the living organisms we know today, it did so through evolution and natural selection but with premeditated planning to allow such things to occur. [As John F. Haught points out in his book "God after Darwin: A Theology of Evolution", the sterile debate between Darwinian evolutionists and Christian apologists is fundamentally misdirected with both sides focusing on an explanation of underlying design and order in the Universe, suggesting that what is lacking in both sides is the notion of novelty. As we will see later, the conditions required for the creation of a Universe that will allow evolution to take place are so staggeringly specific that the odds of this event coming from out of nowhere are mathematically very close to zero (strictly speaking, the odds are not exactly zero, but they come so close to nil that they are practically nonexistent). And these conclusions do not come from any sacred texts; they come from hard facts that modern science is not able to explain away. If, as some critics maintain, religions do not provide satisfactory answers to some of the many unanswered riddles, science is in no better shape either to explain these riddles away.] Since natural selection operates entirely under its own rules, seemingly without any divine intervention whatsoever while it is taking place, the only way in which an act of creation could have paved the way for complex organisms to appear some time after the creation of the universe would have been by preparing the whole scenario from the very beginning in such a way that indeed natural selection would happen sooner or later. And this would send us back to the initial conditions that preceded natural selection itself.
Before going any further, we will take a departure to talk about a game called “Life”, devised back in 1970 by John Horton Conway, a mathematician at Gonville and Caius College of the University of Cambridge. In his book The Recursive Universe that deals extensively with some of the results obtained by carrying out the game of Life, William Poundstone states the following:
“Life is described as a game or, sometimes, a video art form. Neither label quite captures the appeal of Life. Certainly Life is nothing like familiar video games. No one ever wins or loses. Life is more like a video kaleidoscope –a way of producing abstract moving pictures on a television screen … But it’s more than that. The Life screen, or plane, is a world unto itself. It has its own objects, phenomena, and physical laws. It is a window onto an alternate universe … Shimmering forms pulsate, grow, or flicker out of existence. ‘Gliders’ slide across the screen. Life tends to fragment into a ‘constellation’ of scattered geometric forms suggestive of a Miró painting … Much of the intrigue of Life is the suspicion that there are ‘living’ objects in the Life universe. Conway adopted (John) Von Neumann’s reasoning to prove that there are immense Life objects that can reproduce themselves. There is reason to believe that some self-reproducing Life objects could react to their environment, evolve into more complex ‘organisms’, and even become intelligent … Conway wanted to create a game that would be as unpredictable as possible, yet with the simplest possible rules. Conway experimented with many sets of rules. He is said to have devised a game he called Actresses and Bishops. After further thought, he concluded that the rules could be more simplified yet. The simplified game became Life.”
The game of Life is a particular case of a much wider variety of mathematical entities known in technical terms as cellular automata. A cellular automaton is a one-, two- or three-dimensional grid of cells, with each cell representing an independent automaton which in the case of the game of Life can be in one of two states: “dead” or “alive”; and whose next state will depend on the current state of its neighbors. It is dynamic, and starting from an initial condition it will evolve according to a strict set of rules.
The rules for the game of Life are as follows:
Birth: If exactly three “live” cells are neighbors to an empty (“dead”) cell, the empty cell comes to life, and if the cell was already “alive” it will remain so.
Survival: When a cell has two “live” neighbors, it retains its current state (if it was “dead”, it remains dead; if it was “alive” it remains alive).
Overpopulation: When a cell has four or more “live” neighbors, the cell dies from overcrowding.
Underpopulation: Any cell with one or no neighboring “live” cells dies.
These rules were picked out by Conway among many other possible sets of rules in order to satisfy the following design requirements:
- The effect of the neighboring cells in any of the cells upon which the rules are being applied is position independent, and by this we mean that it does not matter where a specific neighboring cell may be touching, whether it is touching just on a corner, above, below, to the left or to the right. Only the quantity of “live” cells actually touching a given cell matters.
- Only the immediate neighboring cells produce an effect when going from one generation to the next. There are no rules incorporating effects from cells that are not immediate neighbors.
- No simple patterns should grow without limit.
- Some simple patterns should be able to grow, provided they eventually reach an upper limit.
- A simple pattern should be able to evolve for a long time before it becomes stable.
A neighboring cell is any cell that touches another cell either along an edge or at a corner.
As an example of how the game evolves, let us start out with a square grid consisting of only 64 cells arranged as eight rows and eight columns. A “live” cell will be represented by a darkening of the cell, whereas a “dead” cell will be represented by a blank cell. Let us begin with the following initial configuration:
Figure 6.1
The reader should take some time to convince himself that if we start with the above initial condition, the pattern corresponding to the second generation will look like this:
Figure 6.2
Take, for example, the “live” cell located in the third row and sixth column of the initial configuration. Since it is being touched (in the lower left-hand corner) by another “live” cell (the one located in the fourth row and fifth column), by the rule of “underpopulation” this cell dies, and when it dies the darkened cell will be replaced by a blank cell in the second generation. Likewise, the “live” cells located in the fourth and fifth rows of the third column of the initial configuration will also die out as a result of the rule of “underpopulation”, since each one only has the other one as its sole neighbor. However, the “dead” cell located in the fourth row and fourth column of the initial configuration will come “alive” by the rule of “birth” because in the initial configuration it had three “live” neighboring cells: the two cells in the fourth and fifth rows of the third column, and the cell in the fourth row of the fifth column, and thus in going from the first generation to the second generation this blank “dead” cell will be replaced by a darkened “live” cell.
The reader should also take some time to verify that the game will evolve into the next three generations as follows:
Figure 6.3
Figure 6.4
Figure 6.5
If we keep applying the same set of rules, the line of three cells now goes into a perpetual cycle, oscillating from the horizontal pattern of the fifth generation to the vertical pattern of the fourth generation and back again. Not surprisingly, this pattern is called a blinker.
There are many other patterns, such as the glider. This pattern moves through the grid by copying itself, and in one of its possible configurations it has the following shape:
Figure 6.6
Among other interesting simple patterns we can mention, capable of evolving into more complex structures, we can cite the T-tetromino (it is made up of four live cells and is shaped like the letter “T”, resembling a similarly arranged stack of dominoes), which will develop into what has been dubbed as the “traffic light pattern” that consists of four symmetrically arranged blinkers:
Figure 6.7
An even more interesting simple pattern is the R-pentomino, made up of five live cells arranged as follows:
Figure 6.8
This amazing simple pattern will evolve into several patterns. At the 48th generation, it will have turned itself into a pattern known as the “Herschel”. As it continues to evolve, it will take other shapes, such as the pattern known as the “Honey Farm”, besides creating and ejecting gliders. It has been found that the R-pentomino will eventually settle down into a final steady state when it has reached the 1103rd generation, filling a low-resolution screen with 25 different Life objects.
There are still many other Life patterns exhibiting “organisms” that may be stationary, periodic, surviving, and disappearing; such as the “barge”, the “snake”, the “mango”, the “hat”, the “fishhook”, the “lake”, the “shillelagh”, the “sinking ship”, the “aircraft carrier”, and so on. There are even patterns such as the “eater” that will do precisely that: eat other patterns they may encounter.
The game of Life is usually implemented and carried out on personal computers; and the grid normally contains some 20,000 or more cells (containing at least some 100 horizontal rows by 200 vertical columns) [many interesting executable programs for John Conway's game of Life are available throughout the Internet (for free!), so if the reader wishes to expand experimentally his knowledge and experience of cellular automata, or is just interested in playing around with this type of technical gadgets, the World Wide Web is a good place where to start hunting for them. Just be sure to search under the topics "game of Life" or "cellular automata"], so there’s plenty of room to create many interesting initial patterns. At this point, the reader may be wondering if there could be other rules we can devise to try to come up with other variants of the game of Life. However, the reader can verify for himself that some of the possible alternatives will lead to nothing of interest, regardless of how the initial patterns are defined to be. To cite just one example, perhaps the simplest, assume that the rules for a new game which we will call the “game of Nothing” are defined as follows:
Birth: If a “dead” cell has one or more neighbors, the cell comes to life.
Survival: If a “live” cell has one or more neighbors, the cell remains alive.
Death: If a “live” cell has no neighbors, the cell dies.
If we start out with a completely blank grid, filled with “dead” cells, it will remain forever blank. Nothing will ever come out of it. If, on the contrary, we start out with a grid completely filled up with “live” cells, the grid will remain forever stagnant. The only way we can extract some action is to start out with at least two “live” cells touching one another. But, alas, this pattern will quickly grow from a small blot in the screen to fill up the grid completely in just a few cycles, with nothing interesting happening “in-between”. No oscillators, no gliders, no nothing. And if we start out with other complex patterns, their fate is sealed even before the game starts, since we know they will eventually and very quickly disappear into a big blot or vanish into a blank grid. We can predict the two possible outcomes with absolute certainty, and we do not even have to set things into motion to foretell the two possible outcomes.
It should now be crystal clear that in order to make it possible for certain complex patterns to evolve into many interesting shapes and situations, the starting patterns are not enough; an adequate set of rules are also needed. Taken in conjunction, the initial patterns and the rules of the game are both the necessary and sufficient initial conditions for an interesting game to evolve. Once the initial pattern (or patterns) on the entire grid has been chosen, everything else will take place entirely on its own, and all we have to do is just sit back and watch as the drama of our creation unfolds.
A much more interesting game of Life would be one in which each cell can take not just one of two possible states but perhaps ten or even a hundred different possible states, with each state being represented by a different color out of the many different possible colors available from a rainbow. Thanks to the widespread availability of fast modern personal computers, this is actually being done throughout the world, and is the subject of intense and active research. Even with all that has been discovered to date regarding these cellular automata, there is the firm conviction among researchers that we have barely scratched the surface. More about this will be said in the next chapter.
An even more interesting variation would be one that, besides allowing for a cell to take many different possible values instead of just two, would also allow us to use a three-dimensional grid instead of the flat two-dimensional grid we have used so far. But then, if we call each cell an “atom”, wouldn’t this start to resemble a game much more familiar to us, the honest-to-goodness true game of everyday LIFE itself?
Again, as in the case of the much more simple game of Life devised by John Conway, we suspect and indeed can quickly verify that the evolution of an interesting three-dimensional multivalued game of Life will depend solely not just upon the initial patterns we choose to put into our three-dimensional grid, but also upon the way in which the rules of the game are written. In such a game, if we use the following rules (to cite an example):
Birth: If a non-colored cell (completely blank) has one or more neighbors with any color, the cell will come “alive” taking at random any of the many possible colors available, regardless of the colors of its neighbors.
Survival: If a colored cell has one or more neighbors that are also colored, then the cell will remain “alive” retaining its original color.
Death: If a colored cell has no colored neighbors, the cell will “die” going blank.
We can verify for ourselves that this apparently more complex game will quickly lead nowhere. The three-dimensional multivalued grid will go completely blank in just one single step, or it will quickly fill up with a stagnant assortment of variegated cells, in spite of the fact that we have introduced an element of chance into the game by allowing a cell coming to life to take at random any of the possible colors available in the game.
The conclusion is inescapable, and we cannot avoid it. The only way in which a cellular automaton such as the game of Life can allow interesting patterns to evolve and interact is by using an adequate set of rules. The rules and the initial patterns are both the necessary and sufficient initial conditions for this to happen, whether we throw in or not an element of chance into the game. For a much more complex game such as this one, it stands to reason that the initial conditions need to be chosen with the utmost care among a myriad of possible choices; otherwise we may quickly end up with stagnant scenarios or dead universes with nothing worthwhile happening “in-between”.
When specifying a cellular automaton, we can take one of two different approaches: either the naturalist approach, or the engineering approach. If we take the naturalist approach, then we are really just experimenting, trying out a specific combination of rules and initial patterns just to see what may come out of it as the patterns begin to evolve, perhaps waiting to see if order will arise out of chaos. A naturalist is primarily looking for patterns that may occur naturally as the cellular automaton evolves. But when we take the engineering approach, we are then trying to build our rules and initial patterns with some specific purpose in mind, we expect that our creation will be equipped to carry out some anticipated actions, most likely complex actions. The naturalist approach requires no more than an observer, whereas the engineering approach requires a designer. When John Conway came up with the game of Life, he took an engineering approach; he very definitely had some specific objectives in mind that he wanted to accomplish. The naturalist approach requires a tinker, whereas the engineering approach requires a clever designer (perhaps extremely clever!).
In his book The Age of Intelligent Machines, Raymond Kurzweil cites a quotation attributed to Robert Wright in a comment to Edward Fredkin’s (former head of MIT’s Laboratory for Computer Science) theory of digital physics (according to the theory of digital physics, the ultimate reality of the world is information processing–or software- and thus this reality should not be described as particles and forces like the atoms and the force of gravity but as bits of data that are being modified constantly in accordance with prescribed computational rules):
“Fredkin…is talking about an interesting characteristic of some computer programs, including many cellular automata: there is no shortcut to finding out what they will do. This indeed, is a basic difference between the ‘analytical’ approach associated with traditional mathematics, including differential equations, and the ‘computational’ approach associated with algorithms. You can predict a future state of a system susceptible to the analytic approach without figuring out what states it will occupy between now and then, but in the case of many cellular automata, you must go through all the intermediate states to find out what the end will be like: there is no way to know the future except to watch it unfold…There is no way to know the answer to some question any faster than what’s going on…Fredkin believes that the universe is very literally a computer that is being used by someone, or something, to solve a problem.”
Grudgingly voicing a similar opinion, Frank Wilczek and Betsy Devine in their book Longing for the Harmonies write:
“We therefore suspect, from its design, that our world just might be an intricate program working itself out on a gigantic computing machine. This form of paranoia may seem extravagant and, of course, doesn’t get to the bottom of explaining the world. We would still need to understand the principles on which the computer was built. For instance, if it is made of silicon, why do the electrons in that silicon obey the laws of physics –are they perhaps fantasies in yet another computer? … Nevertheless, it may not be completely useless to follow up on this suspicion. First of all, it does begin to address the great ‘why’ questions in a rational, if possibly mistaken, way. Second, it suggests a fascinating new sort of question: How would errors in the workings of the computer show up? Could we look for them systematically, by experiments, and thus put the idea of an underlying machine to a scientific test? … Finally, and perhaps most important, thinking along these lines will help prepare us for the day when we –or, more likely, our distant descendants- will develop the machinery and cleverness to begin to program worlds ourselves (and watch –with what feelings? –as the inhabitants of those worlds come to start suspecting …).”
Let us now go back to our discussion of natural selection. Richard Dawkins writes the following near the end of his book The Blind Watchmaker:
“The whole book has been dominated by the idea of chance, by the astronomically long odds against the spontaneous arising of order, complexity and apparent design. We have sought a way of taming chance, of drawing its fangs. ‘Untamed chance’, pure, naked chance, means ordered design springing into existence from nothing, in a single leap. It would be untamed chance if once there was an eye, and then, suddenly, in the twinkling of a generation, an eye appeared, fully fashioned, perfect and whole. This is possible, but the odds against it will keep us busy writing noughts till the end of time. The same applies to the odds against the spontaneous existence of any fully fashioned, perfect and whole beings, including –I see no way of avoiding the conclusion- deities … To ‘tame’ chance means to break down the very improbable into less improbable small components arranged in series. No matter how improbable it is that an X could have arisen from a Y in a single step, it is always possible to conceive of a series of infinitesimally graded intermediates between them. However improbable a large scale change may be, smaller changes are less improbable. And provided we postulate a sufficiently large series of sufficiently finely graded intermediates, we shall be able to derive anything from anything else, without invoking astronomical improbabilities. We are allowed to do this only if there has been sufficient time to fit all the intermediates in. And also if there is a mechanism for guiding each step in some particular direction, otherwise the sequence of steps will career off in an endless random walk … It is the contention of the Darwinian world-view that both these provisos are met, and that slow, gradual, cumulative natural selection is the ultimate explanation of our existence. If there are versions of the evolution theory that deny slow gradualism, and deny the central role of natural selection, they may be true in particular cases. But they cannot be the whole truth, for they deny the very heart of the evolution theory, which gives it the power to dissolve astronomical improbabilities and explain prodigies of apparent miracle.”
Let us assume that, in order to watch our rudimentary “life” forms evolve, we have a digital computer at our disposal and a monitor (preferably a color monitor, though this is not mandatory). We must now ask ourselves again the very important question: Can primitive, rudimentary patterns resembling some of the happenings which take place in ordinary life (such as moving around while preserving shape, competing, coming together, eating neighbors, etc.) be expected to appear spontaneously after the simulation has started? The answer is a definitive YES. If we have already started out with a given set of initial conditions, and if the rules of the game are followed rigorously with no deviation whatsoever, then what seem to be primitive “life forms” may indeed show up in the screen of our monitor or whatever other scenario in which the simulation is taking place.
Simulations that can produce these primitive patterns resembling life forms have sparked the imagination of many computer programmers allowing them to consider the possibility of creating within the bits of software flowing inside a digital computer what has now been dubbed A-Life or “Artificial Life”.
But, wait a minute!
The assumption that out of sheer randomness we can expect to see primitive “life form” patterns evolve in the monitors of our computers is whimsical at best. Earlier we have noted that in order to carry out the computer simulations with some degree of success it is important that the rules which we will now call “natural laws” be followed rigorously throughout the simulation. Any alteration of the rules in the middle of the simulation will most likely ruin our simulated evolution. And if the rules are being followed precisely down to the letter, there is nothing random about the way in which the game is proceeding.
On the other hand, in order to happily watch our rudimentary life forms evolve, the margin of error for each calculation inside our computers must be zero. If just a single bit out of the many thousands or perhaps hundreds of thousands of bits is interpreted as a zero instead of a one or a one instead of a zero (this could happen if a very small portion inside one of the memory chips of the computer is starting to experience electrical failure), then this may be enough to bring our simulation into ruins. The margin of error allowable inside the computer for the correct simulation (or evolution) to take place is zero. This requires a computer that works correctly not just 90% or 99% or perhaps 99.999999% of the time, but 100% of the time. The reliability of the computer must be 100% throughout the simulation. Anything less will just not do. And there is nothing random about this. Reliable machines are designed on purpose to be reliable, and the warranty each machine carries with it is a tribute to this fact. Furthermore, if we take a look at the entrails of our computer, from the major components all the way down to the microscopic integrated circuits, we can bear witness to the fact that there is absolutely nothing random, nothing left to chance about the way in which a computer is built, and everywhere we look inside we find evidence of the knowledge and intelligence possessed by those who designed the building blocks of our computer. Something as sophisticated as a modern personal computer cannot be produced by random chance alone either here or anywhere else in the universe, regardless of how long we may wait for such a thing to happen. And this is the very substrate that we are using to let our “game of Life” evolve! The designers of personal computers, far from being “blind watchmakers”, are for the most part people with college degrees, many of them with advanced degrees in fields such as solid-state physics and electronics engineering.
Likewise, the mathematician or computer programmer who creates on purpose the rules for the game of Life or any other similar cellular automaton game turning those rules into precise steps that will have to be obeyed rigorously by the computer since the simulation begins is no “blind watchmaker” either (as we will see in the next chapter, the selection of the rules out of a vast universe of possibilities will almost certainly demand a good degree of cleverness from the creator). True, he may not be fully certain as to how a particular game will evolve with certain patterns once those initial conditions have been chosen. But at all time he has full control over the experiment, and if he so chooses he can stop the simulation and modify the initial conditions in order to carry out a different simulation, using his free will and his intelligence to carry out his plans. There is absolutely nothing we can find to be random here either.
So much for the supposed “randomness” which the game of Life is assumed by some to have!
So, the answer to the question “Is the watchmaker truly blind?” depends on the extent to which we define the watchmaker. If we exclude all of the initial conditions from the big picture and assume that all of the necessary conditions for evolution to take place (the right temperatures, an adequate atmosphere, enough diversity of chemical elements and compounds to allow in time for complex systems to assemble and evolve out of simpler ones, enough usable energy sources to overcome the second law of thermodynamics and beat large overwhelming odds, etc.) have been handed down gratuitously from out of nowhere, then indeed the watchmaker is as blind and as dumb as it can be; just as in the game of Life where once the existence of a computer or any other equivalent medium in which to carry out the simulation is assumed and the rules have already been given, then the evolution of the game itself will be an automatic process which will inevitably take place even without the presence of any intelligent being to witness it. But if we draw into the big picture the initial conditions themselves, then the only way in which the watchmaker can still be considered to be blind is to assume that the initial conditions themselves are also blind, coming out of nowhere and perhaps rearranging themselves without any help very much as we would expect from any evolutionary process. We have already seen that the very substrate in which an interesting game of Life is played out obeys strict rules which leave nothing to chance; and the machinery itself in which the game runs, far from being a random contraption, bears the mark of an intelligence more advanced and complex than any pattern which might appear and evolve on the game of Life grid. Even a blind watchmaker such as evolution will not be able to assemble a mediocre watch if it doesn’t have any usable parts to begin its work. The real “brains” must be found in the initial conditions that are being prepared for the drama, not in the actual unfolding of the drama itself. If the minimal initial conditions necessary for evolution to take place at some point in time are not there, there will be no evolution, regardless of how many billions or trillions of years have elapsed. Evolution is as dependent upon the initial conditions for its own survival as we are dependant upon the very air we breathe to carry on.
We praise the hardware allowing Windows, Linux and the Internet to run on personal computers as the end result of intelligent engineering, yet many despise the substrate allowing ourselves to run as as the end result of something that according to them cannot even be classed as dumb, a Janus-faced attitude with which many of these skeptics somehow manage to feel comfortable and sure of themselves. Or at least that’s the image they seem to project.
