An interesting excerpt from Collapse by Jared Diamond, about the native intelligence of New Guinea highland farmers, the utility and longevity of that knowledge:
New Guinea is the large island just North of Australia…lying almost on the equator and hence with hot tropical rainforest in the lowlands, but whose rugged interior consists of alternating ridges and valleys culminating in glacier-covered mountains….The terrain ruggedness confined Europeans to the coast and lowland rivers for almost 400 years, during which it became assumed that the interior was forest covered and uninhabited…It was therefore a shock, when airplanes chartered by biologists and miners first flew over the interior in the 1930′s for the pilots to see below them a landscape transformed by millions of people previously unknown to the outside world…..we know now…that agriculture has been going on there for about 7,000 years–one of the world’s longest-running experiments in sustainable agriculture.
…their farming methods are sophisticated, so much so that European agronomists still don’t understand today in some cases the reasons why New Guineans’ methods work and why well-intentioned European farming innovations failed there. For instance, one European agricultural adviser was horrified to notice that a New Guinean sweet potato garden on a steep slope in a wet area had vertical drainage ditches running straight down the slope. He convinced the villagers to correct their awful mistake, and instead to put in drains running horizontally along contours, according to good European practices. Awed by him the villagers re-oriented their drains, with the result that the water built up behind the drains, and in the next heavy rains a landslide carried the entire garden down the slope to the river below. To avoid exactly that outcome, New Guinea farmers long before the arrival of the Europeans learned the virtues of vertical drains under highland rain and soil conditions. (page 280)
My observation is that the knowledge of the European and the New Guinea highlander were different in one very crucial way: the way that the knowledge was acquired. The knowledge that the New Guinea highlander had was acquired over many generations, iteratively, through trial and error. Such knowledge could be called evolved knowledge, because it evolves, good replacing bad, and no overall theoretical framework is required to advance such knowledge. It also had a unique relation to the site at which the knowledge was acquired.
The European agricultural adviser, though had been taught the knowledge, probably in a university. That knowledge was derived from a theoretical model, which made explicit claims to universality. To be sure, properly executed scientific research does make note of the assumptions of its processes and the consequent limitations of its conclusions. However, the realities of the commercial/academic complex place a premium on knowledge which is transferable; in fact the entire network of universities exist to transfer this knowledge. So the limitations and exclusions of scientific inquiries are frequently excluded from application of this knowledge. The aim of the commercial network is to de-particularize the knowledge as much as possible, often through the use of a theoretical model which describes the rule-based behavior of a system. (See the post about CopyBot Resistance which discusses artifacts in which this particularization is important)
Thus, to the extent that the theoretical model is imperfect, or that model’s limitations and exclusions are forgotten, that knowledge is open for mis-use. Through processes of trail and error (release early, release often, from free software culture comes to mind) knowledge that evolves, perhaps even without the need of a theoretical framework, incrementally experiences self correction at each failure point. This suggests evolved knowledge has less catastrophic failures than derived knowledge.
The above insights open a perspective to re-value traditionally acquired and retained information. If that traditionally acquired information is revalued, those institutions which retain and transfer this evolved knowledge can be seen to have a utilitarian value, a value which is perhaps entirely outside of their explicitly stated reasons for existing. Those institutions are examples of persistant networks within society. An example would be a religious institution.
The other interesting insight is that evolved knowledge and derived knowledge also tend to have different failure modes: the derived knowledge tend to be catastrophic, whereas evolved knowledge perhaps tends to be less so. (This is a very provisional speculation)
The areas in which evolved and derived knowledge tend to be useful are interesting also. If I am designing a spaceship (perhaps very expensive that can just be done once) and I am able to view it in isolation (or at least I am able to usefully numericize all the interactions the space ship will have with other systems) then I will use knowledge derived from a theoretical model. If, on the other hand, I am planting a food crop in my backyard, and it would be impractical to gather all of the information about how my garden would interact with the environment (through multiple feed back loops, some poorly understood) then the use of evolved knowledge through persistent social networks makes more sense (or translated to regular English: I might ask my neighbor what worked in his garden, over coffee and doughnuts after Church)
The problem in urban planning that this relates to this is the frequent occurrence that the more trivial criteria (for example number of people that Piazza San Marco can hold) can frequently be numericized with extreme precision, while the harder to quantify variables are also immensely more important (for example, the desirability of creating a space such as Piazza San Marco). The reliance on traditional norms (acquired through the evolutionary processes) when making these decisions has a very deep, authentic rationale.
Often those who would detract from traditional norms use aspersions of: shallowness, or sentimentality when describing such norms. The concept that such norms may be the result of evolutionary processes occurring within society gives a very interesting and much deeper defense than many of the arguments used to defend them to date. The observation that such knowledge may be lacking in a theoretical framework and yet have a more valid process for its creation is an important justification for the retention and further study of traditional norms. (There are parallels with evolutionary computing methods.)