If biology is a science, then ...

Jul 24 2012 Published by under Uncategorized

Since the time of Isaac Newton (the 1600s), scientists have tried to explain the natural phenomena around themselves in terms of other observable natural phenomena. They observe something happen, and then play a "what if?" game to see whether they can explain something else in terms of the first one. That is, they try to extrapolate from something they can study to something they cannot study, by the simple expedient of assuming that the "same thing" is occurring in both cases.

Newton, the apple and the moon

For example, Newton (famously) observed an apple fall from a tree, and concluded that there must be some natural explanation for this behaviour. He had previously observed the behaviour of the moon (assumed to orbit around the earth) and the sun (where the earth orbits around the sun). It then occurred to him that these three observations might all be manifestations of the same "thing", which we now call gravity. All he had to do was assume that gravity acts in the same way on the apple that is right next to him as well as on the stars and planets that are millions of miles away. Thus, by studying the behaviour of objects as they fall on earth, where we can freely perform experiments, we can learn something about the behaviour of stars. From this simple approach Newton derived a whole scheme of celestial mechanics, which made him one of the most famous scientists in history. Not bad for a young man.

Charles Lyell did the same thing for geology. We can observe flowing water washing away soil to form ditches next to roads and paths. This natural phenomenon takes relatively little time to happen. But what happens if we have a lot of time, perhaps millions of years? Well, Lyell suggested that we'd end up with the Grand Canyon. So, once again, an observable phenomenon is used to explain another observable phenomenon — all we have to do is assume that the same "thing" happens over very large distances and/or times.

Aerial view of the Grand Canyon

We can't observe those large distances and times, of course, so the "what if?" exercise cannot be tested experimentally. Nevertheless, this is the basis of modern science — observe and experimentally study whatever phenomena you can, and then use this as a basis for explaining those phenomena that can't be subjected to experiment. If the explanations form a coherent set of ideas, and those ideas are capable of predicting as-yet-unobserved phenomena, or they tie together a lot of observable phenomena, then the "what if?" exercise will be considered to be a success.

Some people reject this approach, of course. These people are not scientists, and there is no reason why they should be. They have their own explanations for observable phenomena, and their own opinions about what needs explaining and what does not. We often refer to these explanations as "super natural", because they postulate the existence of things beyond the observable world. If super-natural phenomena exist, of course, then scientists can say nothing more about them than can anyone else, because science is restricted to a study of natural phenomena only.

The point for this blog post is that in biology we take the scientific approach. For example, we accept that we and the plants and animals around us have a genealogical history. I have a father and a grandfather and a great-grandfather. I have personally met the first one (I talked to him on the phone just the other day) but I never met his father. However, I am sure that his father existed, as did his father in turn, because other people did observe these "phenomena". (What a way to talk about your own relatives!) Furthermore, this observable genealogical history involves relatives that all look different and yet somehow also look the same. Indeed, the closer they are as relatives then the more similar they usually look. You cannot put my father and his two brothers together, along with their several sons (see one of them here), and not suspect that they are all closely related. It's almost embarrassing!

Modern biology is based on the idea that this history of relationships proceeds for millions of years into the past and involves all living things. That is, through time the differences and similarities among all individuals and species have arisen through genealogical descent. In one sense, this is no different from geology and the Grand Canyon — small patterns accumulate through time to create big effects.

There is, however, one difference from geology and physics, and it has created enormous difficulties for the study of biology, not just in the modern world but for centuries. The phenomena being observed seem to be much more complex in biology. As Craig Bohren has pointed out: "The prestige of physics originates partly from its success in achieving its aims. This success, however, has been obtained by applying extremely complicated methods to extremely simple systems ... The electrons in copper may describe complicated trajectories but this complexity pales in comparison with that of an earthworm."

Thus, those scientists studying the non-biological world have often failed to grasp the difficulty of the task being undertaken by biologists, and they have frequently looked on biological experimentation with disdain. What is worse, however, is that it is much harder to explain biology to non-biologists. One only has to look at the plethora of books trying to explain biology to the general public to realize that physicists apparently have it much easier — there are far fewer books because the concepts only need to be explained once.

We now refer to biological complexity as "biodiversity", to distinguish it from the much simpler diversity observed in the non-biological world. I do not know why it took until the 1980s to create a word that we had clearly needed for 3000 years!

The concept of biodiversity leads us to the inevitable question: what does genealogical history look like if we extrapolate it over hundreds of millions of years? It is unlikely to be a simple linear sequence, as in geology, or a mathematical extrapolation from falling apples to orbiting planets, as in physics.

The study of long-term genealogy is called phylogenetics, and the genealogy is called a phylogeny. The latter is a term created by Ernst Haeckel in the late 1800s to describe the reconstructions of species histories that occurred in response to Charles Darwin's ideas on biological evolution. But what does a phylogeny look like?

If we take the observed phenomenon of a genealogy, it is often called a "family tree". Historically, these trees show the male lineage descending from some specified ancestral male, as shown in the picture below, which refers to the current Swedish Royal Family. This shows a branching arrangement, rather than a linear sequence, with (usually) several children in each generation. Nevertheless, a linear sequence does exist in the diagram, since the tree follows only the lineage of those people who became monarch.

The family tree of the House of Bernadotte, the current Swedish Royal Family

(It is worth noting here that recently Sweden change the laws of inheritance, so that it is now the oldest child of either sex who becomes monarch, rather than the oldest male. We thus currently have a "crown princess", Victoria, rather than a "crown prince", as would be usual in other monarchies. This did not happen in the previous generation, for example, where Carl Gustaf became monarch, rather than any of his four older sisters.)

However, a family genealogy isn't a tree either, is it? The different lineages inter-connect through reproduction. My own ancestry, for example, can be traced through either my father's family or my mother's; and their ancestry can be traced through either their own mothers or their fathers.

So, a genealogy is a complex thing. Does this make a family tree a good model for a phylogeny? That is, we are extrapolating from something we can observe, a family genealogy covering several generations, to something we cannot observe, a phylogeny of species covering millions of years. We are also jumping from individual organisms (people) to groups of organisms (species). This has turned out to be more complex than extrapolating from an apple to a planet. This issue is something that phylogeneticists have struggled with for a long time, both in communicating with each other and with non-experts.

I shall look at the possible solutions in the next post.


2 responses so far

  • Jim Thomerson says:

    I think, given the popularity of CSI programs on TV, that we should take a CSI approach in explaining historical sciences to laypeople. The universe, the earth, history of life, etc. are analogous to the crime scene, and, we the investigators, are applying modern scientific methodology to understand and interpret the evidence left by the event. Sir Karl Popper uses the term verisimilitude, truth likeness, to refer to our goals. I recall reading a detective story, where the detective was continuously seeking verisimilitude. It took me a while to understand what he was talking about.

  • brooksphd says:

    Looking forward to the next post!