My usage of 'commensurability' versus Kuhn's
I have now nearly finished The Structure of Scientific Revolutions. During the course of reading that book, I have had many reflections about the notion of commensurability. When I have been using the word to myself in my own thoughts, and in these blog entries, I have noticed that I have been starting to use it in a way that is more and more consistent with the way that Kuhn uses it himself. It would therefore be helpful for me to devise a set of equivalents between Kuhn's language in my own.
I have thought of simply revising my theory of commensurability so that its language follows Kuhn's more closely. But on closer inspection, I find that I cannot. I have worked hard to keep the usage of the word 'commensurability' in line with the dictionary definition of 'measurability by a common standard', and I am not certain that any better word will be found. For when I say that a claim or theory has a certain level of commensurability, I am very definitely referring to its measurability by a very specifically delineated common standard. That standard is its conduciveness to a non-coercive convergence of opinion. In this respect, three different levels of commensurability are very helpful to distinguish. On the other hand, when Kuhn uses the word 'commensurability', he seems to be talking of measurability by the common standard of a paradigm. Hence, merely to claim that two different paradigms are incommensurable is simply to say that one paradigm cannot be measured in terms of the other. It would, in effect, require a meta-paradigm to commesurate the two paradigms--although it seems to me that Kuhn himself provides exactly this meta-paradigm on p.169 of the Third edition of Structure. He says that the unit of scientific achievement is the solved problem, which certainly seems to imply at least in principle a meta-paradigm that employs such a unit. One might, indeed, envisage a utilitarian calculus of maximisation of solved problems--though Kuhn would disagree that this could be the only basis for paradigm choice.
Nevertheless, there is a certain species of discussion where I think that Kuhn's use of 'commensurability' will be clearer. Therefore, when I use the word 'commensurability' without specifying a level, that level will default to level 2, which is most consistent with Kuhn's use of the word. When I use the word 'incommensurability' without specifying a level, the usage will default to level 3 commensurability. Again, I regard that this will be the level that is most consistent with Kuhn's use of the word.
- When Kuhn says 'commensurability', (I think) he means Level 2 commensurability.
- When Kuhn says 'incommensurability', (I think) he means Level 3 commensurability.
I have thought of simply revising my theory of commensurability so that its language follows Kuhn's more closely. But on closer inspection, I find that I cannot. I have worked hard to keep the usage of the word 'commensurability' in line with the dictionary definition of 'measurability by a common standard', and I am not certain that any better word will be found. For when I say that a claim or theory has a certain level of commensurability, I am very definitely referring to its measurability by a very specifically delineated common standard. That standard is its conduciveness to a non-coercive convergence of opinion. In this respect, three different levels of commensurability are very helpful to distinguish. On the other hand, when Kuhn uses the word 'commensurability', he seems to be talking of measurability by the common standard of a paradigm. Hence, merely to claim that two different paradigms are incommensurable is simply to say that one paradigm cannot be measured in terms of the other. It would, in effect, require a meta-paradigm to commesurate the two paradigms--although it seems to me that Kuhn himself provides exactly this meta-paradigm on p.169 of the Third edition of Structure. He says that the unit of scientific achievement is the solved problem, which certainly seems to imply at least in principle a meta-paradigm that employs such a unit. One might, indeed, envisage a utilitarian calculus of maximisation of solved problems--though Kuhn would disagree that this could be the only basis for paradigm choice.
Nevertheless, there is a certain species of discussion where I think that Kuhn's use of 'commensurability' will be clearer. Therefore, when I use the word 'commensurability' without specifying a level, that level will default to level 2, which is most consistent with Kuhn's use of the word. When I use the word 'incommensurability' without specifying a level, the usage will default to level 3 commensurability. Again, I regard that this will be the level that is most consistent with Kuhn's use of the word.
Comments