But to reduce myself to anti-reduction would be terribly reductionist, don't you think? — Moliere

Reductionism has been an extremely successful methodology, but even a reductionist must be puzzled that there are so many branches of science.

Reductionism has been an extremely successful methodology, but even a reductionist must be puzzled that there are so many branches of science. — tom

Some thinkers sympathetic to reductionism, e.g. E.O. Wilson in his Consilience, believe that the divisions between the natural sciences (and perhaps even between the natural sciences, the social sciences, and the humanities) are merely artifacts of our current knowledge base, and that such divisions will eventually fall away as the putative deeper connections between these fields' respective theories become better-understood.

However, in his review of Consilience, Jerry Fodor pointed out that, far from reducing the number of scientific fields, building such theoretical bridges can just as often spawn new fields (e.g. neuroeconomics), which proliferate faster than university deans can keep up with them.

Scientific methodologies have been successful. Theories have explained diverse phenomena.

What need have we of saying 'reduction'?

Physics isn't reductionist. The physical world isn't "just matter in motion", or some such. Not everything is explained by the 2nd law of thermodynamics -- it's not all "just entropy increasing". Neither is chemistry. There are two broad pillars of chemical theory -- thermodynamics and kinetics -- and several subsets of chemistry which focus on the reactions of chemicals in many various ways.

But, then, these statements turn on a particular way of looking at "reductionism". If one just means that some phenomena can be explained by some simpler and more general statements, then OK. But by that same statement I'd say some phenomena can't be explained by some simpler and more general statements -- perhaps they require another simpler, more general statement, or they are an anomaly of sorts.


Ah, @Arkady did a better job than I, I think, just as I was writing this.

Physics isn't reductionist. The physical world isn't "just matter in motion", or some such. Not everything is explained by the 2nd law of thermodynamics -- it's not all "just entropy increasing". — Moliere

Well, a physicist would certainly seek to explain the 2nd Law of Thermodynamics, and in doing so may achieve a reduction or a unification.

Neither is chemistry. There are two broad pillars of chemical theory -- thermodynamics and kinetics -- and several subsets of chemistry which focus on the reactions of chemicals in many various ways. — Moliere

Quantum mechanics is also extremely important, providing as it does the explanation for the existence of atoms, the various phases of matter, atomic bonding ...

I'd say some phenomena can't be explained by some simpler and more general statements -- perhaps they require another simpler, more general statement, or they are an anomaly of sorts. — Moliere

I gave a list above of fundamental objects of certain theories that cannot be reduced: replicators, variation, selection, information, steam-engines, universal computers, perpetual-motion machines of the 2nd kind.

To the above I'm going to add knowledge, which seems to play a fundamental role in an emerging physical theory.

Strictly speaking the the instances of replicators that occur in the Earth's biosphere are genes - portions of DNA that have specific information encoded in them. — tom

Genes or DNA, would it be reductionist to say that they behave the way they do because of chemistry and physics?

In which I did not mention the Turing Machine, which is abstract, but rather the Universal Computer, which is real — tom

The former is is widely known in the literature. Can you give me references for the latter?

Strictly speaking the the instances of replicators that occur in the Earth's biosphere are genes - portions of DNA that have specific information encoded in them. — tom

So DNA are molecules. Genes are portions of DNA. Genes are replicators. Is there anything here you disagree with?

but even a reductionist must be puzzled that there are so many branches of science. — tom

There is a chapter in a book of Steven Weinberg entitled "Two Cheers for Reductionism". Would he count as a reductionist?

In that case, reductionism is simply a mistake and obviously so. — tom

Reductionism has been an extremely successful methodology, — tom

Could you provide a synthesis for our benefit?

Genes or DNA, would it be reductionist to say that they behave the way they do because of chemistry and physics? — Frederick KOH

I'm not concerned so much about resisting reductionism in this case, as being wrong.

Given any particular gene, it can be sequenced. The sequence can be encoded in ASCII or any other format, gzipped, emailed, stored on a USB stick, transmitted, copied, read out loud, etc. From such information, the gene could then be recreated and reinserted into its niche.

A reductionist would have to explain that in terms of the Schrödinger equation.

Perhaps the simplest demonstration of the impossibility of reducing NeoDarwinism to physics, is to recognise that ND does not specify the physics or chemistry required: that Life could exist under different physical conditions or different histories.

Given any particular gene, it can be sequenced. The sequence can be encoded in ASCII or any other format — tom

So what is being encoded is the sequence of molecules cytosine (C), guanine (G), adenine (A), and thymine. We are in agreement here. They are molecules. Would it be reductionist to say that why they and related molecules behave the way they do is because of chemistry and physics?

The former is is widely known in the literature. Can you give me references for the latter? — Frederick KOH

Here it is again:
http://www.daviddeutsch.org.uk/wp-content/deutsch85.pdf

This paper is more accessible:
http://www.daviddeutsch.org.uk/wp-content/ItFromQubit.pdf

A reductionist would have to explain that in terms of the Schrödinger equation. — tom

Actually it has been done.
https://en.wikipedia.org/wiki/Chemical_bond

Genes are portions of DNA — Frederick KOH

...that encode the information required for them to remain instantiated in their niche.

Would be more accurate I think.

Could you provide a synthesis for out benefit? — Frederick KOH

For your benefit, I'll point out the distinction between a methodology and the misconception that higher level explanations cannot be fundamental.

So what is being encoded is the sequence of molecules cytosine (C), guanine (G), adenine (A), and thymine. We are in agreement here. They are molecules. Would it be reductionist to say that why they and related molecules behave the way they do is because of chemistry and physics? — Frederick KOH

It would be wrong. The niche of these genes also includes animal behaviour.

As to what is encoded in the genes, as I have already alluded to, it is information, which you refuse to accept is a proper object of study by the natural sciences.

You could take a step further and recognise that it in fact the information content that is being copied, and it is the information that causes itself to remain instantiated in its niche. In Constructor Theory, this type of information is called knowledge.

Here's an interesting paper: http://constructortheory.org/portfolio/the-constructor-theory-of-life/

Actually it has been done.
https://en.wikipedia.org/wiki/Chemical_bond — Frederick KOH

That's not an explanation of what a gene is and/or of what genes do. Likewise, listing the components of a computer and specifying the way they are soldered together doesn't amount to an explanation of the way such a computer is supposed to function or of what the programs are that such a computer can run. There is much more to a gene being the gene that it is than the nature of the chemical bonds that hold together in DNA molecules the sequence of nucleotides encoding it. The existence of those chemical bonds merely are enabling conditions for those molecules being able to carry stable functional structures from one (or two) living progenitor(s) to its(their) progeny (i.e. whole living organisms). The whole epigenetic context -- which includes many determinate features: (1) of the wider cellular structures and functions; (2) of the whole organism; and (3) of its extended phenotype, and (4) of many aspects of its normal ecological niche (natural affordances) -- plays much more of an explanatory role than do the low level molecular enabling conditions of those high level functions and structures.

The existence of those nuclear bonds merely are enabling conditions for those molecules being able to carry stable functional structures from one (or two) living progenitor(s) to its(their) progeny (i.e. whole living organisms). — Pierre-Normand

They are not nuclear bonds. You don't need to go further than chemical bonds (valency, van der Waals, ionic, etc

The existence of those nuclear bonds merely are enabling conditions for those molecules being able to carry stable functional structures from one (or two) living progenitor(s) to its(their) progeny (i.e. whole living organisms). — Pierre-Normand

Surely you know that DNA replication is something that has been explained at the level of individual molecules. What does "enable" mean in the context of molecules obeying the laws of physics?

Would it be reductionist to say that why they and related molecules behave the way they do is because of chemistry and physics? — Frederick KOH

Does reductionism fail in your view if protein folding via free energy minimisation counts as an NP complete problem? Or is it OK to be hand-wavingly approximate about even these "simplest" computations that nature appears to carry out in holistic fashion. Does it harm your case to admit the sum of the parts are not literally "a sum" when it comes to chemical and physical systems?

Surely you know that DNA replication is something that has been explained at the level of individual molecules. What does "enable" mean in the context of molecules obeying the laws of physics? — Frederick KOH

DNA replication is one thing, genetic inheritance is another. The inheritance at issue is inheritance of function. Biological function only can be explained with reference to the high level functional organization typical of specific forms of life. If you abstract away from the context that gives significance to physiological processes, then you are doing physics and chemistry all right, but you have given up providing a biological explanation. You have just narrowed the focus to questions of material constitution, which are just one sort of question one can ask about a biological system.

For your benefit, I'll point out the distinction between a methodology and the misconception that higher level explanations cannot be fundamental. — tom

In the example that you gave, they were fundamental only because they were studies of abstractions, of idealized objects.

One of the things that tends to happen is an object of study inspires an abstraction. Then the abstraction becomes a field of study in its own right. It is in this context that they become fundamental,

Geometry is an abstraction inspired by solid objects in space. It is fundamental. But being an abstraction it does not concern itself with questions like why metallic spheres don't collapse into each other when they are pressed against each other. That answer is provided by quantum mechanics.

if protein folding via free energy minimisation counts as an NP complete problem? — apokrisis

Modelling physical phenomena using mathematics? How novel!

Or is it OK to be hand-wavingly approximate about even these "simplest" computations that nature appears to carry out in holistic fashion. It doesn't harm your case to admit that the sum of the parts is not literally just "a sum" when it comes to chemical and physical systems? — apokrisis

Have I said anything to suggest otherwise?

Have I said anything to suggest otherwise? — Frederick KOH

You tell me. I'm unclear whether you are simply defending reductionism on the grounds of epistemic utility or - as it does sound - trying to make a strained ontic claim.

DNA replication is one thing, genetic inheritance is another. — Pierre-Normand

Molecules in motion is one thing. Pressure, temperature and volume is another.

All you had to do was quote a comment of mine.

Biological function only can be explained with reference to the high level functional organization typical of specific forms of life. If you abstract away from the context that gives significance to physiological processes, then you are doing physics and chemistry all right, but you have given up providing a biological explanation. You have just narrowed the focus to questions of material constitution, which are just one sort of question one can ask about a biological system. — Pierre-Normand

I will let Steven Weinberg answer this one:

When Edelman says that a person cannot be reduced to molecu-
lar interactions, is he saying anything different (except in degree)
than a botanist or a meteorologist who says that a rose or a thun-
derstorm cannot be reduced to molecular interactions? It may or
may not be silly to pursue reductionist programs of research on
complicated systems that are strongly conditioned by history, like
brains or roses or thunderstorms. What is never silly is the per-
spective, provided by reductionism, that apart from historical ac-
cidents these things ultimately are the way they are because of the
fundamental principles of physics.

All you had to do was quote a comment of mine. — Frederick KOH

All I've asked you is whether it matters that protein folding can't be completely modelled as an addition of local bonding forces. Surely you accept that as proof that "something" goes missing once one tries to reduce the rate-dependent dynamics of the real physical world to a rate-independent informational description?

Surely you accept that as proof that "something" goes missing once one tries to reduce the rate-dependent dynamics of the real physical world to a rate-independent informational description? — apokrisis

Isn't the very idea of abstraction leaving things out?