Turnbull, D. (1989). Maps are Territories : Science is an Atlas. Chicago: University of Chicago Press. (Reprint edition 1994).
"In these two passages Michael Polanyi and Thomas S. Kuhn equate theories with maps, and they take it for granted that the metaphor is self-explanatory. Indeed, the map metaphor is not only used to describe scientific theories, but is so pervasive that it is also commonly employed to illuminate other basic but ill-defined terms such as `culture', `language' and `the mind' (see ITEM i.l). Since metaphors play a very important role in science and in all our thinking about the world, we should be alert to the depths of constructed and construable meaning contained within them (see Putting nature in order, pp. 54-7). It is particularly important in this instance because there is no clear understanding amongst scientists, philosophers or cartographers as to what either a theory or a map is. " (Turnbull, 1989, p. 1).
"Exhibit 8
MAPS-A WAY OF ORDERING KNOWLEDGE
Karl Popper argues that languages have a descriptive function which is clearly distinguishable from their argumentative function. This, according to Popper,
makes the familiar analogy between maps and scientific theories a particularly unfortunate one. Theories are essentially argumentative systems of statements: their main point is that they explain deductively. Maps are non-argumentative. Of course every theory is also descriptive, like a map.
Karl Popper, Unended quest: an intellectual autobiography, 1976, p. 77)
Popper may be mistaken about maps, possibly because he is concerned at this point to make a distinction between the descriptive and argumentative functions of language. Though the distinction can be sustained analytically, it cannot be pushed too far because there is a powerful sense in which descriptions are arguments. The strongest sense in which that is true is illustrated by maps. They invariably carry less information about the environment than is out there, since they are necessarily selective, but they also frequently carry more information than was actually recorded. Some maps, like those used for navigation, can have data plotted on them and deductions as to position and distance to destinations made from them. But the example of Wegener's the continental drift (see ITEM 8.1) [omitted] and the cholera map (ITEM 8.2) [omitted], show that neew knowledge can be gained from a map in a profound and significant way. It was Wegener's 'fitting' of the continents that led him to hypothesise the original joining of South Africa and South America. It is interesting that, while Wegener used that `fit' as evidence for his theory, his opponents were able to reverse the argument and criticise his theory on the grounds that map-fitting did not constitute evidence and that the apparent fit was the result of selective distortion. Mapping the outbreaks of cholera also reveales distribution that could not have been simply read from the data. Moreover maps are not always mere re-presentations of data. In ITEM 8.3 [omitted] the magnetic stripes in rocks on the sea floor off the coast of West Canada become in their turn evidence of the phenomena of sea-floor spreading from ocean ridges, the ultimate proof of Wegener's theory. Indeed ordering information spatially provides a very powerful mode of inference to knowledge that is ignored by Popper.
However, there is an even deeper sense in which maps are like theories in that they are 'argumentative systems of statements'. That is the sense in which they embody or express a cognitive schema. As has been discussed in Imagining nature, observation statements are not clearly separable from theoretical statements, and theoretical statements in turn embody sets of assumptions about how reality is ordered. This whole complex of unconscious assumptions about the ordering of reality which structures our experience of it can be thought of as a cognitive schema. In the case of maps, the idea that our ability to understand the world is dependent on modes of ordering of which we are at best only partially aware is of particular interest. As concrete example they provide an opportunity for bringing such cognitive schemas to the fore, and they also provide an opportunity to explore the claim that in the deepest possible way is inherently spatial, and embedded in practical action." (Turnbull, 1989, p. 48).