The Triple Helix: Gene, Organism and Environment [Paperback]

In the first movement (Gene and Organism), Lewontin reviews major discoveries in biology from Darwin to the Genome Project. In his critique the author carps the metaphors of biology, especially the once useful words and phrases like Decarte's metaphor of the world as a "machine", general use of the word "development" (unrolling or unfolding of something that is already there) to mean ontogeny and embryo genesis and the "Holy Grail", i.e., the Genome Project (the project that determined the nucleotide sequence of the entire human genome). Using elegant examples from contemporary biology, Lewontin dispenses with the ideas (1) that a cell is anything much like a machine and (2) that as a blueprint, DNA sequencing would be sufficient to define anatomy, development and function.

In the second movement (Organism and Environment), the author clears up the meaning of "ecological niche". Accordingly, environment and organism are so closely related that, except in the laboratory, neither exists in the absence of the other.

"Organisms not only determine what aspects of the outside world are relevant to them by peculiarities of their shape and metabolism, but they actively construct, in the literal sense of the word, a world around themselves."

In movement three "Parts and Wholes, Causes and Effects"; the reader is treated to a glimpse into Lewontin's home life:

"As I write this chapter I think at one moment of the sentence I am writing, but then I wonder which sonata my wife will practice next, and then I recall the work done by the plumber today and then I return my attention to the manuscript."

Also included in movement three are (1) highly instructive lessons on values of fitness of nine genotypes in the Australian grasshopper (2) a discussion of variety among ceratopsian dinosaur horns and collars (3) a story about a Vermont man with a 150 year-old axe, and (4) the history of infectious disease in nineteenth century Europe.

In the finale, Lewinton dispenses with holism, Gaia, catastrophe, chaos and complexity theories and adds:

"Rather than searching for radically different ways of studying organisms or for new laws of nature that will be manifest in living beings, what biology needs to do to fulfill its program of understanding and manipulation is to take seriously what we already know to be true . . . the fact that biological systems occupy a different region of the space of physical relations than do simpler physico-chemical systems . . ."

and

"New experimental techniques are in part induced by the problems that are under investigation by a community of scientists with common interests, but once those technologies exist they have great power in determining the questions that are asked."

The same is said of great men.

In this lecture Professor Lewontin outlines the role that genes, environment and chance ("random noise") play in the development of an organism. As he phrases it on page 20: "the organism is not specified by its genes, but is a unique outcome of an ontogenetic process that is contingent on the sequence of environments in which it occurs." This means that you could take the same genetic code and have it unfurl in Hyde Park and get an organism different from one you would get having it unfurl on, say, the Boston commons. Lewontin shows how cuttings from the same plant cultured at different altitudes developed differentially, and in a manner that could not be predicted. The reason they could not be predicted is that there is a significant amount of random variation ("developmental noise") that occurs as the plant grows. Lewontin gives the further example of a multiplying bacterium on page 37. The bacterium divides in 63 minutes. In another 63 minutes the daughter cells should divide again, giving four bacteria, but actually there is some random variation in how long it takes them to divide, so that one daughter divides in say 55 minutes, the other in an hour and five minutes. And this continues so that the bacteria culture does not increase in pulses, but continuously in random increments. This difference in timing in multi-cellar organisms may result in morphological differences since a catalytic enzyme may arrive too late to, say, grow a side bristle on a fruit fly (an example that Lewontin gives). Lewontin applies this understanding to the development of our brains on page 38. First there are random connections set. "Those connections that are reinforced from external inputs during neural development are stabilized, while the others decay and disappear." This process, Lewontin advises us, can lead to differences in cognitive function that are neither strictly genetic nor strictly environmental. They are influenced by random (unpredictable) factors.

This understanding is the reason that Lewontin is less than thrilled with the Human Genome Project. He believes, as he makes clear in another book, It Ain't Necessary So: The Dream of the Human Genome and Other Illusions (2000), that we will be disappointed by what can be accomplished simply from sequencing the genetic code, his point being that even though we know the code, the environmental and random factors cannot be known in any precise or predictive sense. It is true that the genome for a chimp will always code for a chimp and never for a rabbit, but whether that chimp is good at math or has unusually aggressive tendencies is something we cannot know from an understanding of the genetic code alone. Chance and environmental factors in development can result in a passive chimp even though its parents are aggressive.

Applying this idea to evolution in general, we can see that individual variation is not strictly a result of environmental differences but also of chance differences. Consequently, what we are is not shaped strictly by adaptive pressure (natural selection) but is to some extent the result of purely random processes. At one time in my life I studied chance and random events, and one of the most important things I learned is that the term "random" is not clearly defined, except in the sense that something that is random is unpredictable, which is a "you can't prove a negative" sort of definition. I also learned that there is considerable doubt as to whether a truly randomizing device actually exists. All real world devices, such as roulette wheels and computer random number algorithms can be shown to have some tiny bias, or to break down at the extremes. (Don't trust the random number generator on your computer when you are generating a very large number of trials: it will begin to repeat, and your Monte Carlo simulation will be flawed.) So what Lewontin calls "random events" are actually events that we simply do not know enough about to describe accurately. It may be that with greater ability we will eventually be able to describe or control these events. However, it may also be that at some level such events are the direct result of the probabilistic nature of a quantum event, and therefore in principle unpredictable. I suspect that Lewontin believes something like this.

In the second lecture Lewontin makes the point that to a significant degree organisms create their environment, and it is wrong to think of a place (such as the surface of the moon) without organisms as an environment. His dictum is "...[T]here are no environments without organisms" (p. 67). In the third lecture Lewontin discusses some of the problems associated with genetic causation and its analysis. There is a fourth chapter in which Lewontin attempts to provide some direction for future studies in biology.

I did not understand his assertion on page 81 that "Only a quasi-religious commitment to the belief that everything in the world has a purpose would lead us to provide a functional explanation for fingerprint ridges or eyebrows or the patches of hair on men's chests." The hair, I imagine is the result of sexual selection, but surely the fingerprint ridges allow us a better grip, and our eyebrows shade the sunlight as well as providing some small cushioning for our eye sockets.