“The ‘tool versus science’ debate has received little mention in the published literature of geography “ (p. 348), despite the attention given to GIS. A non-issue for most practicioners and researchers? Wouldn’t it fall within the scope of any ongoing debate on epistemology and methodology (philosophy of science) that researchers and the scientific community must bring to bear on any chosen method? Shall we assign the term “science” to any methodology, or is the methodology itself one component, albeit an important one, of scientific inquiry? Would a more fruitful framing of the debate involve the appropriate application of GIS-as-a-tool to the type of research in question, again, as with any chosen method? I guess I’m confused about the nature of the debate, and whether there really is much of a debate here beyond a handful of researchers beating their fists on an open door. (In a time of viral online phenomena and 1000s of hits instantaneously, the description of 64 comments among 40 people over 34 days as an intense discussion seems quaint and severely overstated.) The authors of the article appear to have their hat in the ring on the GIS-as-science end of the spectrum. I can see that the design of GIS, as with any research instrument, can be approached scientifically and empirically, but I am not aware of a theory of GIS (I readily acknowledge my ignorance here!). I tend to agree that geography or other disciplines are the science, GIS the tool.
I agree with Wright’s comment that “GIS encompasses the way in which geographical info. is collected, perceived, managed, and used” (p. 351), but this point would apply to any scientific methodology and not cause us to redefine the methodology as a science. My sense is that doing GIS is part of doing science, but not a science unto itself. The point of Goodchild (1994) that GIS can be applied to the spectrum of scientific approaches including positivist and nonpositivist approaches to me supports the view that GIS is more appropriately understood as a tool rather than a science.
Is the question driven by a now-outdated concern that because GIS labs, courses, and GIS-based research require significant investments of resources, that such investments would be more likely if given status of science unto itself? And perhaps even a desire for greater legitimacy? It seems now that the wide acceptance of GIS as a tool has helped address both these concerns. The Goodchild update “Fifteen years on” may provide some insight.
Mabu