Smith College professor, Robert Dorit, has written a very fine reflection on the persistent problem of reductionism - the illusion that by breaking something into it's parts you have fully explained it. Despite increasing awareness within the most reductionist disciplines - biology, chemistry and physics - that such is not the case, significant institutional and cultural momentum over the previous centuries means this thinking is still alive and well.
Dorit explains that reductionist thinking is a very significant problem in science and culture because it represents a hazardous blind spot. In essence, we've become highly skilled at taking things apart but have much less skill in synthesizing and integrating that knowledge.
He points out that we know a great deal about neural functions but are still unable to explain consciousness with similar detail. We know a lot about genes but don't really understand the most significant aspects of the relational web that stands between genes and a school of fish that migrates to find food. Complexity science and, I would argue, network science, may represent a turn away from reductionism and toward integration.
Analysing something by breaking it into smaller and smaller parts has yielded great insight and we shouldn't disparage that. What we need to be very cautious of is any claim that such reductive analysis represents complete knowledge. It clearly does not. Reductive analysis becomes reductionism when there is a blind commitment to disassembly as the final form of knowledge. We must avoid that and call it out when we see such assertions made at the expense of integration and synthesis.
* A tip of the hat to Ben Ramalingam from Aid at the Edge of Chaos. It was his post that pointed me to the Dorit article and is well worth visiting for any number of other posts that explore complexity, aid, and problem solving.
Hard-won research specialties are not easily surrendered or modified. That means that we have inculturated reductionism through institutionalizing it. That has it's place, too. It is not, however, a universal model or one that can work in isolation. I think that one of the reasons it is difficult to attract new people into science stems in part from the embedded reductionist paradigm.
Meaning is not usually found buried at the end of an obscure, hyper-specialized piece of knowledge. While that knowledge may in fact be important, it is meaningless apart from a relationship with all kinds of other knowledge. We need institutions and practices that recognize this hazard and that develop new ways of thinking about what we know.
He points out that we know a great deal about neural functions but are still unable to explain consciousness with similar detail. We know a lot about genes but don't really understand the most significant aspects of the relational web that stands between genes and a school of fish that migrates to find food. Complexity science and, I would argue, network science, may represent a turn away from reductionism and toward integration.
Analysing something by breaking it into smaller and smaller parts has yielded great insight and we shouldn't disparage that. What we need to be very cautious of is any claim that such reductive analysis represents complete knowledge. It clearly does not. Reductive analysis becomes reductionism when there is a blind commitment to disassembly as the final form of knowledge. We must avoid that and call it out when we see such assertions made at the expense of integration and synthesis.
* A tip of the hat to Ben Ramalingam from Aid at the Edge of Chaos. It was his post that pointed me to the Dorit article and is well worth visiting for any number of other posts that explore complexity, aid, and problem solving.
Hard-won research specialties are not easily surrendered or modified. That means that we have inculturated reductionism through institutionalizing it. That has it's place, too. It is not, however, a universal model or one that can work in isolation. I think that one of the reasons it is difficult to attract new people into science stems in part from the embedded reductionist paradigm.
Meaning is not usually found buried at the end of an obscure, hyper-specialized piece of knowledge. While that knowledge may in fact be important, it is meaningless apart from a relationship with all kinds of other knowledge. We need institutions and practices that recognize this hazard and that develop new ways of thinking about what we know.
1 comment:
Awesome!
Post a Comment