10 Words About Evolution

Yet again, Gene Expression has an interesting post, this time setting a challenge to describe evolution in 10 words or less. Razib says,

If you had 10 words or less, what would you have the public master (and I mean internalize, not spit back as a creed) about evolutionary theory?

Razib's attempt is this:

Differential fitness correlated with heritable variation results in evolution.

Interestingly, he says that his attempt is meant to condense what he says are Lewontin's "'three conditions for biological evolution: (i) there must be phenotypic variation; (ii) the different phenotypic variants must be associated with different fitnesses; (iii) fitness must be heritable."

Razib's challenge is interesting, and it generated a number of other candidates and some discussion on other blogs, notably at Evolgen and John Hawks' blogs. I'll get to those momentarily. I first want to pick a nit with Razib. He says that he's attempting to condense Lewontin's three conditions for "biological evolution," but those conditions aren't for evolution, they're for natural selection. Here's what Lewontin says in the paper where he sets out the conditions, i.e., "The Units of Selection" (from Annual Review of Ecology and Systematics, 1970, p. 1):

The principle of natural selection as the motive force for evolution was framed by Darwin in terms of a "struggle for existence" on the part of organisms living in a finite and risky environment. The logical skeleton of his argument, however, turns out to be a powerful predictive system for changes at all levels of biological organization. As seen by present-day evolutions, Darwin's scheme embodies three principles:

1. Different individuals in a population have different morphologies, physiologies, and behaviors (phenotype variation).
2. Different phenotypes have different rates of survival and reproduction in different environments (differential fitness).
3. There is a correlation between parents and offspring in the contribution of each to future generations (fitness is heritable).

These three principles embody the principle of evolution by natural selection. While they hold, a population will undergo natural selection.

See a problem? Lewontin is talking about evolution by natural selection, not just evolution. So Razib's try is incomplete because, after all, natural selection and evolution aren't the same thing. RPM, of Evolgen, pointed this out immediately in comments on Gene Expression, but also here. He says:

It seems like most of the readers in the ScienceBlogs universe don't understand the difference between evolution and natural selection. When Razib asked them to provide the thing they would want the public to understand about evolution (in ten words or less, mind you), most of them provided explanations of natural selection. Let me say it as clearly as I possibly can, evolution is more than natural selection.

Absolutely right. But here's a trick we can do with Lewontin's and, so Razib's, conditions. As I argued on this blog a while back, here and here (which generated a nice discussion with John Hawks), Lewontin's conditions don't capture evolution by natural selection. In particular, they don't distinguish between natural selection and drift. Now, we want to say, then, that Lewontin is wrong about the conditions for evolution by natural selection. But we can also be creative and say that Lewontin, and so Razib, is on to something about evolution.

Now, Razib's conception of evolution doesn't define evolution. Rather, it tells us what results in evolution. But he doesn't tell us what accounts for the result, he only tells us conditions. RPM saw this, too. His conception is this:

Differential inheritance of genetic variation via stochastic and deterministic forces.

I like RPM's answer to Razib's challenge because it actually tells us what evolution is rather than setting conditions for it. John Hawks says that

No evolution means equal offspring for everyone!

To be sure, Hawks has a pretty good slogan with his offering. But it doesn't tell us what evolution is; it tells us only what happens without it.

So, who's right? Razib? RPM? Hawks? I think RPM has the best shot. Actually, I'd shorten it:

Differential heritable variation via stochastic and deterministic forces.

Eight words. I cheated, of course, off RPM's paper. (Well, I guess I cheated off everybody!)

Comments

[coturnix]

In the comments on Evolgen, I changed RPM's definition by just one word:

"Differential inheritance of developmental variation via stochastic and deterministic forces"

And the inheritance can be genetic, or epigenetic, or behavioral or whatnot....

[RPM]

I tried to avoid the term 'heritable' as it carries with it connotations of heritability with means something different that most people think. I'm still bothered by inheritance, however, as it seems to limit our discussion to the transmission of alleles from parents to offspring. We need a term that encapsulates both allelic transmission and differential survival (both due to stochastic and deterministic forces).

"Differential transmission of alleles and survival of individuals due to stochastic and deterministic forces."

Darn! That's four words too many.

[Matt Dunn]

I didn't read the comments over at Gene Expression, but all these attempts seem to be attempts to distill what evolution (or natural selection) is, not what you'd want the public to "internalize" about evolutionary theory, as the original challenge was put. I would imagine that the 10 word distillations above are a bit too high-level for the general public. They are interesting puzzles in conceptual analysis for biologists and philosophers, but they don't really seem to convey what evolution is in any sort of vivid and clear way to people that don't have much background in evolution.

I'm very sympathetic to the claim that evolution is more than natural selection. In fact, I've been worried recently that natural selection isn't actually anything at all (to be polemical of course, but really, it is weird to think about what natural selection actually is). In my opinion, the most important thing to convey to the public, and the most wonderful thing about evolutionary theory in general, is common descent and the evidence for it, morphological, behavioral, and, of course, molecular evidence. It is awe inspiring (in some sense) and it has important implications for how we understand ourselves as a species and how we live as a society (of course making this latter point is skating on thin ice, but I believe it can be made and must be made in a responsible way, e.g. Frans de Waal draws some interesting lessons from it I think at least with his bonobo stuff as do animal rights activists, but I don't agree with them all).

My mantra to teach the public and to have them actually understand the evidence for would be "Every living thing is related."

Of course this may leave room for many different types of processes that produce the particular pattern of relatedness we have on earth (ahem, intelligent design), but once people actually understand the evidence for common descent, I mean really understand it, I think they'd be hard pressed to disagree with a theory that posits only naturalistic processes. In fact, in many ways, accepting the evidence for common descent requires accepting a fairly common interpretation of the evolutionary process.

And I don't mean there is the "fact" of evolution and the "theory" of evolution and common descent is the "fact". There is a lot of theory that goes into studies of common descent, into phylogenetics, that is independent of, say, population genetics (population genetics being the "theory" that explains how the process of evolution works). My only point here is that there is also lots of population genetics in phylogenetics, particularly in coalescent models for example.

Of course I'm not so sure how clear and vivid the coalescent would be to the public either. But "Every living thing is related" sure has a romantic ring to it.

[John I]

"The fit tend to live and breed, the unfit don't"

[Roberta Millstein]

I'd avoid the use of the term "deterministic" in this context. People mean very different things by it. For a philosopher, it tends to imply either an ontological claim about the world (something along the lines of: given the complete state of the universe of one point in time, there is only one possible future state) or an epistemological claim about what one would know if one were ominiscient (Laplace's demon could predict any future event with perfect accuracy). It is at least questionable whether either of these philosophers' definitions is appropriate for natural selection (I take it that natural selection is one of the things that one is trying to capture by "deterministic").

Biologists, on the other hand, tend to use the term "deterministic" to say something about the nature of a theory -- namely, that it is a theory that predicts one unique outcome rather than, say, a probability distribution of possible outcomes. This too is inappropriate, because you're trying to define the evolutionary process, not say something about the nature of the theories that people use to try to capture it.

The rule of thumb is: avoid the use of the term "deterministic" if it all possible. (Don't even get me started about the term "stochastic").

Maybe I'm missing something (alas, I didn't read the other blogs), but what's wrong with Darwin's "descent with modification"? Three words.

[Robert Skipper]

Maybe I'm missing something....

I think Dunn's comments on this blog are accurate. We all read the question, and then myopically focused on defining evolution instead. Nevertheless, as much as I like "descent with modification," I --and surely others-- wanted to get as much detail in the definition as we could in 10 words or less.

The rule of thumb....

Hm. It sounds like you've thought some about this. ;)

[Roberta Millstein]

I think Dunn's comments on this blog are accurate. We all read the question, and then myopically focused on defining evolution instead.

Fair point. As much as I like "descent with modification," it isn't obvious what it means unless you think about it a bit.

I --and surely others-- wanted to get as much detail in the definition as we could in 10 words or less.

Well, yeah -- if you're going to use all 10 words. :-)

What would be nice if somehow our 10 words could incorporate the idea of a tree of life. I haven't come up with a way to do that that is clear.

Then again, maybe we just need to resist slogans. Look at what happened with "survival of the fittest"...

Hm. It sounds like you've thought some about this. ;)

Naw, what gives you that impression? :-)