Because most philosophies that frown on reproduction don't survive.

Wednesday, October 12, 2005

Evolution, Speciation and Nominalism

When terms like 'unscientific' start getting thrown around in the various debates about evolution, someone invariably says, "I have no problem with micro-evolution. We observe that all the time. My problem is with macro-evolution. We've never seen one species evolve into another, and I see no evidence that we ever will."

Now, first of all, there is some evidence out there of speciation. However, it's not what doubters generally want to see. When someone asks for evidence of macroevolution, they generally don't mean a new strain of bacteria, a new variation of the fruit fly theme, or a series of slightly different trilobite fossils. They want "sexy science". Something big and dramatic. How about an ape evolving into a hominid?

There are two built in challenges to showing evidence of vertebrate evolution. Producing it in the lab is difficult because vertebrates reproduce on a much, much longer cycle than bacteria or fruit flies. If it takes (to pick a number at random) at least one million generations to produce enough genetic changes that you could possibly have a new species, then it's a lot easier with three generations a day, or an hour, than it is with one generation every six months or every year.

Thus, vertebrate speciation would clearly take a lot of time. And with time, comes difficulty in confirming if it really is speciation anyway, which is where nominalism comes into the equation.

Philosophically speaking, nominalism is a very problematic rabbit hole indeed. If 'good' is just a name which we assign to a variety of things which contain some degree of 'goodness' and yet there is not objective standard of goodness to which we compare each individual instance of 'good' then there is really no such thing as 'goodness' at all, just a handy word we use to label things we like. Nominalism, in its purest form, leads to complete relativism, and as such orthodox Catholics avoid it like the plague.

Now let me be clear, I am very much a realist of Plato's school when it comes to the ideal forms like Beauty, Good, Justice, etc. These universals must have an ideal form to exist at all. However, things become more problematic when you deal with a concept like 'cat' or 'horse'.

I think we must say that there is some sort of spec for what a horse is meant to be. If a horse is born with a missing tendon such that he cannot stand, there is clearly something wrong with the horse. That horse is deviating from how a horse is meant to be, and as such we can say that the horse has a birth defect. So don't put me down as a species relativist. It seems like at any given point, with any given spieces, there is a basic design of 'horse' or 'cat' which members either adhere to or deviate from (Thus, a cat with three legs is a deviation.) However, there can still be variations in regards to accidental characteristics within the type that are not necessarily deviations -- so for instance a white horse and a black horse are both equally 'horse' and neither deviates from the type -- they merely have difference accidents. Other characteristics may constitute a deviation (anyone know what the correct philosophical term would be?) from the type. So, a horse born with a withered leg has a deviation from a type (horses are meant to have four functional legs) but is not because of that less a horse. He remains a horse, but with a deviation from type or defect.

However, when we start looking back into the history of the horse, we start to run into problems. The definition of a species is: "An animal which can produce fertile offspring with members of the same species." So any given horse is a member of the species Equus caballus because he is able to mate and produce fertile offspring with other members of the species Equus caballus, while a donkey is not, because if he mates with a horse the offspring will be an infertile mule.

Now, it's easy to perform these experiments in the present day: All you need is an amorous horse and a number of other hooved animals to try him against and eventually you can work out what is a horse and what isn't. You cannot, however, hop in a time machine and try the horse against various ancestor to see at what point 'speciation' might have occurred. So instead, paleontologists make species distinctions based upon body structure. (I wonder what a paleontologist would make of a great dane vs. a chihuahua. They are not only members of the same species, but of the same sub species, so I assume you could quite successfully mate them -- though I can't imagine either would approve of the idea...)

But what if you could hop in a time machine with your horse and go back in time to try his luck? Well, if evolutionary theory is correct, you'd run into a confusing phenomenon. Something like this would happen: you'd go back 100,000 years, and have no problem mating the horse and producing fertile offspring. Another 100,000 years, and another fertile offspring. Now go back a full 500,000 years into the past. Things are getting hit and miss now. Go back one million years, and sure enough you're producing infertile offspring between your modern horse and the horse population of 1million BC.

However, now you go back and pick up the horse from 500,000 years ago (who successfully mated with your modern horse and produced fertile offspring) and get it together with the 1million BC horse, and you'd probably get fertile offspring. The genetic differences between your modern horse and the 1mil BC horse would sufficient to prevent fertile offspring, but the differences between the 500k BC horse and the 1mil BC horse would be less and might well allow fertile offspring. So you'd have a peculiar situation on your hands. By most basic definition of species, the 1mil BC horse is of a different species than the modern horse. And yet the 500k BC horse is the same species as both the modern horse and the 1mil BC horse. Further, while the 500k BC horse might not be able to produce fertile offspring with a 1.5mil BC horse, the 1mil BC horse probably could. And so on. The fertile offspring definition of species simply doesn't work as far as trying to figure out where exactly the boundary of speciation is going back in time.

Even assuming that the theory of evolution is correct, producing and measuring speciation events is a tricky thing.


Kevin J. Jones said...

Have you read Etienne Gilson's From Aristotle to Darwin and Back Again? I'm told it deals with the problems of speciation and stability amid change over time, among other things.

Darwin said...

Hmmm. I'll have to have a look for it. I read Gilson's "God and Philosophy" and was very impressed with it.

Darwin said...

Wow. That is one hard to get hold of book. Not only does Amazon not have it, but Advanced Book Exchance featured only two copies, both in French and over $30 a piece. Hmmm. This will have to go on the long term search list.

Kevin J. Jones said...

Yeah, it's hard to find. I'm hoping to borrow it from a university library soon.

Steven said...

Dear Darwin,

Problem is, your definition with species to start with is problematic. Rana pipiens the common east-coast frog forms a cline along which there is a degree of intermingling--that is the frogs furthest south can produce fertile offspring with frogs somewhat further north, but those furthest south cannot mate and produce fertile offspring with those furthest north. In addition, there are the ring clines of the Galapagos finches. These all call into question our notion of species.

Island biogeography and allopatric speciation also throw certain wrenches into the works.

The reality is that the concept of species is a very mutable thing, and very dependent upon the taxonomist making the determination. We've long known that there are "lumpers" and splitters. Lumpers will claim that splitter take sexual dimorphism and describe new species on the basis of it. Splitters will say that lumpers need a degree of separation that isn't evident in the biological world to make a splilt.

Personally, I have no problem with microevolution, which in the proper realm of things is the development of new species from old. I do have major problems with macroevolution, which in the field refers to the development of new gnera, families, or classes from a species. My problem with this is probably more semantic than it is real--but ultimately a species can ony give rise to another species. Biologically all higher classes only exist in an abstract reality which we codify. Hence, a species gives rise to a new species that we RECOGNIZE as a new genus or class, but which really is merely a new variant of the old species.

The concept of species is so subtle and so mutable that it becomes very difficult to base a lot on it because the definition is so uncertain. A horse with a withered limb is a horse. There is no genetic difference. But it is not a mule, fertile or otherwise. A tigon is neither a lion or a tiger (and is, in fact an abomination caused by human interference).

Oh well, too much on too little. I really like this blog.



Darwin said...

Indeed, the 'problem' is partly that the animals themselves do not know what species they've been assigned to. They're just doing their thing, being born, eating, having children adn dying. From that point of view, it's all very simple...

I'll have to read up on the frogs you mention. I've heard about a pair of species of butterflies in a similar condition -- in that butterflies from opposite ends of the geographical area cannot produce offspring together, but ones in the middle can, almost as if we're seeing geographical speciation in the process of taking place.

Thanks for reading!

Hans Georg Lundahl said...

I would like you to read the message labelled karyogrammata on this no, not this, but index page of my creationist blog.