Reductionism

I am a reductionist, but in a modest way. I see many put-downs of reductionism which ascribe ideas to reductionism that I have never heard espoused. To me reductionism has always meant that the behavior of the whole stems from the behavior of its parts and the lawful interactions between these parts (thanks Ramachandran). This does not mean that to understand the behavior of the whole it suffices to understand the behavior of the parts. It is neither necessary nor sufficient to understand the behavior of a computer in order to understand the behavior of a program running therein; indeed that is largely irrelevant. Yet the behavior of the computer while obeying that program does indeed depend on the behavior of the computer and the parts of the program represented by the state of parts of the computer.

No one supposes that understanding chess is merely to know its rules. If our brains were without capacity limits then perhaps we might dispense with higher level ‘understanding’ but that is a silly counterfactual.

There are ample reasons to understand things in our world in a non-reductionist manner:

We may not have an adequate theory of the parts while we have a working theory of higher level behavior.
We may not have adequate information storage capacity in or heads or even our computers to hold the data necessary to anticipate the behavior of the whole, even such behavior as is obvious to a child.
We may not have the compute capacity in our heads or even our computers to foresee the future consequences of the current state.
We may not have the means to acquire that state even if we had the capacity to store and compute it.

The above impediments apply even when when we manage to avoid the combined dark forces of quantum indeterminacy, chaotic behavior and lack of infinitely precise measurement tools, such as in understanding behavior of computers and certain other digital systems.

With all of the above caveats one asks if there is anything left of reductionism. I think so. The physical theories that serve us very well are constructed along reductionist lines; especially those the engineers use to build and program computers. There is ample room for theories of behavior of systems not based on behavior of their parts. I would hope that such would be a temporary situation, but many such theories are useful despite such status. Thermodynamics was only belatedly founded on Newtonian mechanics and it served well before it was so founded.

Some suspect limitations of our brains to hold adequate theories of some systems that we aspire to understand. This has been suggested for understanding both particle physics and brains. I am agnostic on this point but lean towards optimism.

Robert Laughlin is a rare physicist who may not be a reductionist even in my loose sense of the word. I think he claims that particles may behave in ways that stem from the behavior of the whole. I shall have to see a coherent mathematical theory along these lines before I even know what this would mean.