Drift, evolution and all that

Continuing the discussion from How Does Drift Contribute to Adaptive Evolution?:

How could drift not contribute to evolution? But “drift” is the wrong term to use here – I’ll come back to that later.

I think I’ll make this a new topic.

I’ll use part of Newtonian science as an analogy.

The Newtonians defined length. But they were basing it on traditional ways of measuring length with a portable measuring rod. They set up a platinum rod in Paris as an international standard for the unit of length.

Let’s imagine what would have happened if they had not set that standard. Maybe we would just use our own measuring rods or rulers. And as the old rulers wore out, we would create new ones. But without a standard, the new ones might be a tad different in length from the old ones. As this process repeated, we might expect the unit of length to change with each new generation of rulers. It would change only slightly. You might call that drift. But, again, “drift” is the wrong term here.

At a time when there was relatively little international travel, we might expect that the “drift” in France would be a bit different from the “drift” in England and a bit different from the “drift” in Russia.

You get “drift” because there is no standard. But actually, there was a kind of standard. The ruler that was being used would be a kind of local standard – a standard for the population that shares its ruler technology. We can think of that as an internal standard. There would be “drift” because of the lack of an external standard.

Philosophy of science

A note about science is appropriate here. The Newtonians were smart enough to understand the importance of standards. So they set international standards for just about everything that they measured.

Here’s my puzzle about philosophy of science. I’ve read a number of books on philosophy of science. And I don’t recall seeing any mention of the importance of standards. To me, this seems a serious flaw in philosophy. The philosophers of science apparently want to concentrate on things like laws of physics. Apparently, coming up with justified true beliefs is fundamental to them. But this seems to miss the point of how science works.

Back to drift

Let me mention some examples of the kind of “drift” I am talking about. An obvious example is the divergence of languages. This “drift” may start as dialect. And as the dialects get further apart, we begin to recognize them as separate languages.

Another example is currency (money). There is no external standard for the value of the unit of currency. So the value will “drift”. We call this “inflation”. In principle, we could see both inflation and deflation. But we mostly see inflation, because the politically powerful find that more acceptable, so they manage to bias the “drift” toward inflation.

What’s wrong with “drift”?

For my example of measuring length, it looks as if the unit of length will drift. But it only looks that way to us, because we are used to thinking in terms of external standards. If we were part of a population where the unit of length were drifting, we would not notice any drift at all. Everything would be in accordance with the only standard that we would know.

Maybe we could say that there would be change over time. But again, we think of it as change only because we think in terms of external standards. I’m inclined to think that the proper term is “evolution” rather than “drift”.


Getting back to science, for the moment, I’ll note that we don’t really have external standards. We have international standards. But the standards of science are not external to the human population. They all depend on human choices.

The Newtonian scientists did their best to provide standards. So they used the freezing point of water as a temperature standard. And they used a platinum rod as a standard of length. These standards weren’t really external, but they did anchor our internal standards to some reasonably stable aspects of reality.

In the case of an evolving (or “drifting”) population, natural selection tends to anchor the internal “standards” of the population to what will allow that population to thrive. So the internal “standards” are anchored to the environment, and will “drift” and the environment changes. And that “drift” will be such as to tend to keep the population well adapted to the “drifting” environment.

A final comment

The usual account of evolution is in terms of genes and genetic change. That sort of account is sometimes said to be “bottom up”. I think I have just given an outline of a “top down” account of how evolution works.


I don’t see where you came back to that. What am I missing, and what is the right term?

I find your account confused. This “standard” you refer to is natural selection, and it doesn’t apply to most of your genome. Populations don’t drift; that would happen only if there were no selection. Some parts of genomes are evolving through genetic drift, while a few are evolving (or are kept from evolving) by selection.

I had a whole paragraph (or two) under the heading ‘What’s wrong with “drift”?’

Ah, I see now. But “evolution” already has a meaning, and it includes both selection and drift. Your substitution is a very bad idea. Then again, I don’t understand your argument at all. Genetic drift just refers to stochastic change in allele frequencies. There’s nothing about a standard or even lack of a standard.

Drift means motion. And motion has to be relative to something.

You say that there is drift is to imply that there is something relative to which it is drifting. So there is an implicit direction, contrary to the usually stated view that evolution has no direction.

1 Like

The motion is a change in allele frequency from one generation to the next.

I am clearly not understanding your point. Does a random walk have a direction?


I would not say so. But I also would not say that there is drift.

I agree with that. I don’t currently have a better way of explaining it.

Is this perhaps an argument over the difference between the meaning of a term in different fields? I seem to recall some old mention that in physics drift refers to directional motion. But in biology it doesn’t. Are you complaining because the same word can mean different things depending on context?

Crow and Kimura, in their 1970 book, noted that. They tried to keep things clear by always using “random genetic drift” instead of just “genetic drift”.

No. It has more to do with my study of human cognition and consciousness. I find that people are unable to understand consciousness, primarily because they are insisting on examining it through their own biases. So I try to avoid that.

If I think in terms of a population, then members of that population won’t see any drift. The appearance of drift comes from how we are looking at it.

You would be wrong on that account. The standards of measurements is a foundation of modern science and philosophy of science.

I think you will be delighted to hear that the scientific community has moved away from arbitrary standards and moved towards standards that can be derived from first principles. The meter is no longer defined by a stick. A meter is the length a photon travels in 1,299,792,458th of a second. The second is defined by the oscillation of a cesium atom. The kilogram is defined by Plank’s constant. If you were flown to a distant planet you could replicate all of these standard units using the physical constants around you.

As best I can tell, there could not be such standards.

But that could only be the oscillation of a cesium atom on earth. The choice of earth is pragmatically convenient, but still arbitrary.

The oscillation of the cesium atom is the same in all frames of reference. It’s governed by physical constants. If you can build an atomic clock you can measure the second.

The speed of light in a vacuum is the same in all reference frames.

Plank’s constant is the same throughout the universe.

These and other constants are now used to define standard units in science. We no longer use a platinum rod or platinum weight.

We don’t know that. And we cannot know that.

The cosmological redshift could be evidence of a different timing. It is assumed not to be. But we do not and cannot know that.

At this point I have no clue about what you’re trying to say or how to respond. Was this an experiment in the response to random stimuli?

1 Like

To be pedantic, the meter was never defined by a stick. It was defined (around 1790) as one/ten-millionth of the distance from the equator to the north pole along a meridian through Paris.

The stick was merely the physical representation of this length.


As much as we know there are physical constants, yes we do know that.

Based on what? We already know that a difference in relative velocity shifts the frequency of light without there needing to be a “different timing”.

I don’t find value in throwing out well supported conclusions because somebody can dream up a fantasy that isn’t instantly disproven.

Nobody has actually measured that difference in relative velocity. It is assumed that the red shift is due to difference in velocity, but we do not actually know that.

Let’s recall my point – that all of our standards are locally based (i.e. human standards). It seems to me that you are wanting to throw out that well supported conclusion.

Are you pulling my leg?
Let me google that for you

That’s wrong. They are now based on universal constants.

No, but perhaps you are trying to pull mine.

Nobody has used Doppler radar to measure the velocity of distant galaxies.