249
Framework, and exemplified in reference to cases of both physical and psychosocial
properties. The Framework was also used to highlight once again the importance of
evaluating the quality of the information produced by a measurement, now described
in terms of the high-level, complementary requirements of object relatedness
(“objectivity”) and subject independence (“intersubjectivity”). Finally, the
Framework provided a sufficient condition for measurability: a property is measurable if it is the input of at least one process that has been successfully structured
according to the Framework.
On this basis we may come back to our opening question: Given the necessary
conditions discussed in the first stages of our exploration and the conclusions
reached in the subsequent stages, what complementary sufficient conditions can we
propose for characterizing measurement across the sciences?
8.3 Can there be one meaning of “measurement” across
the sciences?
As we have already discussed in this book, the International Vocabulary of
Metrology (VIM; JCGM, 2012), possibly the most authoritative existing source
regarding fundamental concepts in measurement and their related terms, gives the
following definition of : “process of experimentally obtaining one
or more quantity values that can reasonably be attributed to a quantity”. Our arguments have pointed that this definition could be revised in three aspects:
• Given the diversity of cases of what an experiment can be (e.g., thought experiments, mathematical experiments), the condition that measurement is “experimental” is too weak: as discussed in Sect. 2.2.1, a measurement should be
constrained first of all to be an empirical process, so as to avoid the conclusion
that purely computational processes can be considered measurements; complementarily, as introduced in Sect. 2.3 and developed in Chap. 7, measurement
must include also an informational component, which produces its result in the
form of values of the measurand.
• What characterizes a property as a quantity is not so clear, as discussed in Sect.
6.5, and, apart from the traditional view that superposes and
, which we have criticized in Sect. 3.4.2, nothing clearly requires a
property to be quantitative in order to be measurable.
• Finally, the concept is too generic, given that it encompasses both
general properties and individual properties, as we discussed in particular in
Sect. 6.1. Individual properties can be identified both as properties of objects and
values of properties, as summarized in Sect. 5.1.3; hence a more explicit reference
is appropriate to the condition that what is measured is a property of an object.
Hence our VIM-like, updated definition of could be something like:
8.3 Can there be one meaning of “measurement” across the sciences?
Précédent

- 282/319

Suivant