Stephen Jay Gould's Full House:
The Spread of Excellence
from Plato to Darwin

(Hardcover, 244 pages. Harmony Books, September 1996. ISBN: 0517703947)

Review and commentary
Andreas Ramos
November, 1997

Introductory Comments

Stephen Jay Gould, professor of geology at Harvard, is very popular in America. Press reviews of him are very favorable. Gould is considered as An Important Scientist. Journalists regard him as "Mr. Evolution." He's a Harvard professor who gives funny talks that include cartoons, jokes, baseball stories, and a winking dismissal of those "stuffy scientists who are short-sighted, but we know better." Gould does a good show.

But at the same time, one reads comments from published, recognized evolutionary theorists who say things such as "It's regrettable that American intellectuals are learning their evolution theory from Gould...", "...it's a pity that he has acheived this recognition, but says nothing..." or that "Gould is a science writer, but I say that through clenched teeth." Gould, a geologist, is not taken seriously by the experts in evolutionary theory. His theories of evolution are ignored by the experts. He's little known outside the USA.

Since the only reviews on the net of his book seem to be promotional advertising, and since Gould considers his book FULL HOUSE is the final presentation of his central argument, it's worth taking a close look at his argument. Let's ask two questions:

• What is Gould really saying in this book? Is this a theory of evolution? What kind of argument is he making?
• What are the implications of his ideas? So what? If he is right, what does it mean for other projects?

Part One: Progress Is Just an Illusion

In part one, Gould makes the point that evolution is NOT directed towards a goal. He repeats over and over that it's a fallacy to think that evolution LEADS to more complex life forms or that it STRIVES towards intelligent life.

I emphasize the words "leads" and "strives": Gould is precisely refuting any teleological or inherent nature in biology. He repeats several times that bacteria are the most widespread and successful adaptation: they have been around since the beginning of life and will be here probably until the sun explodes.

Gould writes:

"I am asking my readers finally and truly to cash out the deepest meaning of the Darwinian revolution and to view natural reality as composed of varying individuals in populations--that is, to understand variation itself as irreducible, as "real" in the sense of "what the world is made of." (p.3, bottom)

Thus part one's refutation of Platonic forms and statistical averages: those tend to see the world as abstract or ideal forms which are the real and to downplay the individual as just a point on a graph. Gould sees it the other way: the individual is the real, the ideal is just a statistical description.

Since the purpose of this review is to investigate the implications of Gould's concept, what does Gould's Darwinian revolution have to say about ethics, justice, art, and various other pursuits? Will Gould take up his own challenge? Will Gould be a Darwinist about ethics, the arts, politics, etc.?

Actually, no. He will never say anything about any of this, neither in this book or in any of his public articles, because for Gould, these are just trivia pursuits of a trivial species. He will argue that bacteria are the major form of life on this planet and that there's never really any change at all.

Part Two: Dead Horses Tell No Tails

Part two is a historical account of the changes in the understanding of the evolution of horses.

Evolutionists at first thought that horses had evolved on the Asian continent (there were horses in Asia but no horses in the Americas). Furthermore, this evolution of the horse was thought to be ladder-shaped, or a series of steps: first a small horse, about the size of a cat, with many toes, later larger, with fewer toes, and yet later, much larger with only one toe. Later it was realized that the real story was much more complex: horses originated in the Americas: their evolution took place over a much longer period, and there were many, many more branches. The neat image of a progressive step-by-step evolution of the horse dissolved into dozens and dozens of forms. Instead of a ladder, we have an articulated bush, of which the modern horse is just a lucky survivor.

What Gould is saying is true, but this has little to do with evolution theory. He is stating a historical account. In fact, it's a particular kind of historical account: it's an account of the changes in historical ideas. Gould is carrying out a postmodernist or deconstructionist analysis of history.

Part 3: Better Batter

Part three fills fifty-five pages, or a quarter of the book. In this major section, Gould comes alive. That's an understatement. In the rest of the book, he plods along, occassionally reaching into a sample jar to pick up a hairy worm and patiently explain how it cleverly adapted to some odd climate. His voice is calm, citing dusty academic journals, pooh-poohing other theories, taking the conservative position, saying "well, we can't really say for certain". Several times he dismisses others by saying that the implications of their ideas are "chilling" (his word.)

But when Gould gets to part three, he wakes up. For, you see, Part 3 is about baseball. Gould isn't an evolutionist. He's a baseball fanatic. Gould would much rather talk about baseball than anything else.

Now for him to have spent his life in measuring snails is indeed scientific work. But... to giggle about charts that show where people stood around in a baseball field? (P. 115) To chortle over the creation of statistical analysis of baseball? Let's quote him...

"Please pardon a bit of crowing, but I was stunned and delighted (beyond all measure) by the elegance and clarity of this result. I knew from my previous analysis what the general pattern would show, but I never dreamed that the decline of variation would be so regular, so devoid of exception or anomaly for even a single year, so unvarying that we could even pick out such subtleties as a deceleration in decline. I have spent my entire professional career studying such statistical distributions, and I know how rarely one obtains such clean results in better-behaved data of controlled expirements or natural growth systems. (...) But the decline of standard deviations for batting averages is so regular that the pattern of Figure 16 looks like a plot for a law of nature." (P. 107-8)

This isn't an exception. The whole chapter goes on like that. Gould in his excess is torpedoing his own argument. He rejects the concept of progress as fuzzy-minded species-centrism for people who don't appreciate statistics. He starts by writing:

"Since 0.400 hitting is both so noticeable and so justly celebrated..." (p. 79)

Now, Gould wants to point out that 0.400 has gone extinct. He wants to show that baseball has evolved according to laws which hold for both biology and baseball so that 0.400 is just a form of centrism whose passing is natural. But... it's clear that he'd much rather notice and justly celebrate 0.400 hitting and anything that has to do with baseball (which form the most thrilling moments of his life, p. 131).

Gould is baseball-centric to an extreme. If he's talking about evolution, he explains Latin in plain English. But with baseball, not once does he bother to explain what he means by "0.400 hitting". I haven't the slightest idea what the number means. A former sports journalist pointed out to me that it's odd that Gould writes "0.400"; the standard form is ".400". One begans to realize that Gould is not a reliable expert. Reading Gould, one might think that baseball is a national religion. Basketball is the main American spectator sport. Baseball has difficulty attracting people to stadiums, even with free seats and free entertainment. If he doesn't report that correctly, one wonders about other things that he says.

Gould turns his private obsession with baseball into a proof of evolution theory (see p. 217-8). If he means it as a joke, well, it's a weak joke. If he means this seriously, it makes him look goofy.

Part 4: Bully for Bacteria!

Part Four is easy to summarize: Gould spent his childhood avoiding his mom's demand that he clean his bedroom.

Gould uses bacteria to illustrates his point that there is neither progress nor zenith in evolution. Those who point to the complexity of man are refuted by Gould who points out that this argument arbitrarily focuses on a minor, recent lifeform. Bacteria have been here longer and there's more of them. Bacteria live in odd places, including possibly up to 2 miles under the earth. By weight, bacteria make up most of the biological mass of the earth. They've been here forever, and they're not leaving. Since bacteria are able to live under the earth and get energy from the planetary core, it's possible that bacteria may live on other planets as well, protected underground from solar radiation, space chill, etc. Thus bacteria may be the dominant form of life in the universe. To think that mammals are proof of progress in evolution is mere mammal-centricity which silences the voices of bacteria.

Pages 168-175 summarize his argument in seven points, for those who just want to skim the book.

Are humans any better than bacteria? Gould concludes that we are mere random events of an unpredictable process with no drive to complexity (p. 216).

In the closing pages, he throws in a few points about culture, but maybe his editor should have deleted this section. These are weak arguments.

Finally, Gould writes that his position isn't some form of political correctness (p. 229). Indeed, he writes that his position of random evolutionary permutations should teach us to celebrate all forms of life. Even the lowly bacteria sings its song. Yeah, right... hand me the Lysol. Sorry, Gould. Your position is bio-political correctness. Now clean your room.

Gould's Conclusions

Gould proved two things: the extinction of 0.400 hitting in baseball and the absence of an inherent trend to complexity (p. 217), both caused by the same law of random statistics.

Gould's first main point is that concentrating on mammals at the expense of bacteria is a bias of mammal-centric observers. Certainly there are more bacteria (ten trillion in a cubic centimeter of dirt), but so what? To dismiss mammals as higher forms by saying that bacteria outnumber us is, well, not much of an argument. What about virus? There's more of those. Is Gould being bactero-centric to favor bacteria over virus? What about prions (protein strands that can cause certain diseases)? What about electrons? There's plenty of those too. Is it bio-centrism to favor bacteria over electrons? If one finds a peak in a statistical collection, that peak can be flattened by looking at a larger set. But if one starts the game of "whatever you say is irrelevant because I can find a larger set", then one can reply with an ever yet larger set (virus, or electrons, etc.) Now, this only proves something about the peak in relation to the entire set, but it doesn't say much about the peak itself: even if there are indeed one hundred trillion billion bacteria, mammals do exist. Gould engages in a fallacy by attacking an argument by shifting the definition of the argument.

Gould uses the example of 0.400 hitting to show that emphasis of a trait is done at the expense of ignoring the larger, general trend. Humans are a flash in comparison to 4 billion years of bacteria. But his own example goes against his argument. Gould is saying that we shouldn't celebrate baseball and proves this by crowing about baseball. This is a bit like the preacher who sermons that he is not one of those preachers who talks about tits and cleavage, no, he won't do that, he won't talk about long-legged blondes and their sex appetites, oh, no! Gould denounces those who concentrate on humans and thus ignore the broader picture. And he proves this with an obsession on baseball. He won't emphasize that certain things are more significant than others, oh, no.

Gould's second main point is that "evolutionary progress" is a bias of human-centric observers. Gould brings up the Drunkard's Walk to illustrate that there is no progress in a statistically random set. This is a well-known statistical illustration: a drunkard who walks randomly in directions will end up walking mostly around the same spot.

Gould argues that evolution is a random permutation of life forms. For example, any species adapts itself to local weather conditions. In cooler times, the species gains hair; in warmer times, it loses hair. At the end of ten thousand generations, there's no net progress, just a random back and forth. For a while, it had hair, for a while, it didn't. If we flip a coin ten million times, we may find that for 100 flips, the coin landed heads, but in general, we'll find that there is no preference. By looking at horses, we find that there was no clear progression from the cat-like horse to the modern race horse; instead, there were all sorts of sizes, variations, and regressions.

Gould is forgetting that there's a bit of a difference between random coin flips and life forms. If we flip a coin, we all know that it's 50/50 each time we flip BECAUSE THE COIN HAS NO MEMORY. Each flip is an entirely new event. The coin is unaware of the last flip. Gould is correct about trends in coin toss because there is no inherent bias towards one or the other side.

But a mouse isn't a coin. Each generation of mice has a history, namely, the history of all previous mouse generations. As a mouse colony encounters a new virus, many will die. But some survive because of genetic variance within the species. The next generation of mice will be made up of survivor mice who are resistant to that virus and other mice which luckily avoided exposure to the virus. Fifty generations later, the mice encounter the same virus again. This time, fewer die because the survivor genes have spread throughout the colony. A mouse is the cumulative result of the experience of every prior mouse generation. We have in our bodies hundreds of thousands of systems that arose in response to events over millions of years, not just as humans, but also as mammals and prior to that as well, down to the single-cell state in the early ocean. Many of these systems are obsolete and serve no purpose, but they remain in our body. Each new generation is able to adapt to the world slightly better, adding new biological skills to the repertoire. Thus each new adaptation adds to the species, resulting in more complex decendants. The random coin flip argument and the Drunkard's Walk illustration do not apply to biological systems.

Gould is claiming that it's mammal-centrism to think that mammals are superior life forms to bacteria. Bacteria have been around for some 3.5 billion years, have distinct forms, etc, but asides from simple reproduction, what can they do? Cats have complex emotional lives, are aware of themselves, communicate with each other, and can be contented with sitting in the sun and watching the yard. Cats build nests to rearrange their environment to suit themselves or simply move somewhere else. All of this is beyond the capabilities of bacteria. Bully for bacteria? Bacteria can't even play catch.

Gould's argument is shortsighted because he is doing statistics, an area of mathematics, instead of genetics, biology, or any physical science. Evolution theory is based on genetics, which is based on changes at the cellular molecular level. Life is an effect of molecular structures. Other researchers in evolution theory and evolutionary epistemology are proposing that life is an effect of quantum effects. For example, it appears that organic compounds, instead of being rare in the universe, are actually quite common: clouds of interstellar dust which can be lightyears in lenght are made up of organic molecules. One of the characteristics of atoms is that they tend to form stable structures, such as crystals. Organic molecules will form into strings. These strings can break apart and then each string will replicate itself. Thus it is the nature of organic molecules to reproduce. If they find an abundance supply of appropriate atoms or organic molecules, they will reproduce.

A complex organic molecule sorts out other molecules, moves to them to the proper location, or rejects the wrong kind of molecules. A complex organic construct, such as any single-celled organism, is a molecular machine that interacts with its environment. The molecular machine can be considered as a sort of computer. That environmental interaction will occur according to a set of rules which are specified by its atomic and molecular nature, including quantum effects. Those rules form its algorithm for interacting with its environment. The purpose of the algorithm is to create a model of its environment and then test that model against the real world. The algorithm processes a model of the organism's environment. It tests against that model.

This can be seen in single-cell organisms. They do not have any sort of nervous system or thought. Their actions are the result of built-in structures. Simply considered as machines, they move around in their environment, avoiding situations which are dangerous and moving towards situations which permit reproduction. All of this can be illustrated with the paramecium, a single-celled organism which you can find in any brackish pond water or aquarium and observe with a simple microscope. It moves about with 5000 to 15000 independently motorized cilia, which are whip-like hairs. These beat and the paramecium moves forwards and backwards. It eats bacteria. It senses stimuli such as mechanical deformation, gravity, chemical compounds, light, or temperature. Ellen Taylor's web page describes paramecium. More details can be found at D . R. Caprette's page. Paramecium have no nervous system or any sort of intelligence, yet interact with their enviroment. It has an algorithm that has an ideal picture of its environment and it is able to manipulate that environment.

Now let's consider the effects of evolution on the world-modeling algorithm. The members within a population of organic molecules, or organic machines, are similar but not identical. Even within a single population, such as paramecium in an aqaurium, there are differences at the molecular level. The population is made up of a range of organisms. As this population of organic machines encounters new situations, some die and others survive. Some will die and some will survive. The algorithms of the survivors will spread among the next generation. Over millions of generations and situations, the survivors will have more complex algorithms.

Since they are rearranging their environment to suit themselves, they are aware of themselves as the agent and goal of their actions. Due to selective pressure, eventually self-aware minds will arise. Nearly all mammals are self-aware. They perceive situations and know whether they are in a play-situation or a hostile situation. They are aware of the mind-state of other mammals and know who is friends, enemies, and so on.

The more that an algorithm can model and re-arrange its environment (e.g., nest building, shelters, etc.), the more it is able to free itself from local environmental constraints. It may seem strange, but animals were among the earliest life forms. Animals arose some 600 million years ago. Plants arose only 400 million years ago (200 million years after animals.) Animals manipulate their environment. Humans have reached the point where they can live in hostile climates, such as Antartica and outer space. Beavers build dams in order to create ponds for their food, housing, and to raise their pups.

Gould concludes that there is no tendancy towards complexity in organisms (based on his statistical idea that a handful of mammals is irrelevant in the face of trillions of bacteria) and that there is no such thing as progress (considered over billions of years, mammals are just a fluke). This results from his reliance on statistics, but as I have shown in the preceeding, Gould is ignoring physics, molecular chemistry, and biology. Ellen Taylor's web page on paramecium is part of her high school science project, but armed with a 100x microscope and her careful observations, she realizes that there is a remarkable complexity in paramecium, which is apparently is beyond the capabilities of Stephen Jay Gould, professor of geology at Harvard.

Due to the nature of the atomic stucture of the universe, there will be:

• organic molecules
• organic molecules will reproduce
• organic molecules will act as biological machines
• biological machines will tend towards complexity
• biological machines as algorithms will be aware of their environment
• biological machines will be aware of each other and thus form social environments
• biological machines will be self-aware

The universe may well be filled not only with life, but also intelligent, self-aware life that lives in cultures. Organisms such as mammals, which are able to control and manipulate their environment, really are a higher form of life than bacteria and there really is progress towards self-awareness.

I'll point out that at the moment, general evolution theory does not believe that there is a tendency towards complexity or progress. It is my suggestion that complexity and progress arise due to considering organic molecules as biological machines with an epistemological algorithm that assists in coping with the environment.

For those who are interested, here is an excellent summary of current evolution theory. It gives a much better account of the relationship between bacteria, animals, and plants and their evolution than Gould's baseball games.

General Comments on Gould's Project

Gould has managed to place himself in the American spotlight as Mr. Evolution. He glorifies baseball and uses it to illustrate a minor law of statistics about random events. This point can be stated briefly: if we flip a coin 100 times, we may see lucky streaks of five heads, but in general, those lucky streaks are meaningless. That in essense is Gould's theory of evolution.

Gould thinks that life forms, incl. mammals and humans, are mere random events. He has no theorectical basis for discussing the selfish gene, the sharing gene, the cultural gene, or anything that doesn't play baseball. No wonder he has little to say about the IQ debate or any other genetic ethical issue. It may sound odd to say this about "Mr. Evolution", but Stephen Jay Gould is actually anti-evolution theory. His book is a long argument that evolution doesn't really happen.

To put it very plainly: Gould does not believe that evolution happens. In his book, he talks about statistics, but not genetics. For him, species are random permutations. If mammals should occur and lead to humans, well, let's take the long view of 10 billion years and we'll see that humans are just a tiny blip in an ocean of bacteria. For Gould, there is no evolving, no development, no direction, no improvement. There's no need to discuss whether culture is a genetic trait. No need to look at language as a species-specific ability. No bother whether IQ is genetically determined. It's a waste to time to study genes, DNA, etc. Bacteria are the dominant form of life and they are completely stable. His "Darwinian challenge" turns out to be empty: to any position, he'll dismiss it as a random event. Gould dismisses others by saying that they are "privileging a position" and "erasing the margin". Those concepts come from postmodernism and political correctness. Those arguments are not biology, evolution, nor science.

Gould's rejection of change has a serious consequence. In the Dec. 1999 New Yorker, Robert Wright points out that Gould's argument is used by Creationists to argue that evolution can't produce higher life forms and humans and therefore there is a god. Gould is used by anti-evolutionists to argue against evolution.

Others have pointed out that Gould is not doing evolution at all. Evolution theory is based on genetics and the changes that occur at the molecular level, including discussions of quantum effects (see e.g., Roger Penrose or David Deutsch.) Gould doesn't discuss any of this. Instead, he looks at the change in understanding that has occurred through the history of evolution theory. He's not looking at evolution theory itself; he's looking at the changes in perception. This is epistemology (the theory of human understanding). Gould points out the first understanding of horse evolution and then the second understanding of horse evolution. Gould points out the proper understanding of statistics. Gould points out the mammal-centric bias and shows that bacteria are more significant. All of this is an attempt to control and change the definitions. That is a strategy from political correctness and French postmodernism.

Gould doesn't say much but he says it well. Gould's reputation is based on popular books that are amusing, witty, and full of references to pop culture. He politely supplies explanations of the tough words so that everyone can follow along. His baseball comes across as a little boy's charming cleverness to turn an adult discussion into a discussion of his pet hobby. His books are not science and they're not really popular science either. It's entertainment.

There won't be a "Gould School of Thought" (loosely stated: "Bacteria are the dominant form of life and in the long run, nothing really happens.") There's not much more to add after that.

In closing, I'd like to mention those snapdragons in the illustrations at pps. 11-13: isn't that the same fish? Twisting leap, snapping away? Same body position, same place in the drawings, etc. I think someone borrowed from someone.

footnotes...

• Harvard's Evolution web page
• UC Berkeley's Evolution web page
• Selected Bibliography
• Donald Campbell

Is this Page Useful? Vote!

Follow   Me!

Updates: andreas.com newsletter