103
mately caused by or supervene upon low-level cognitive and neurophysiological
processes.
This realism does not necessarily entail a naïve correspondence view of truth that
neglects the role of models in knowledge, nor does it deny that our conceptual
schemes, models, and linguistic frameworks actively shape our experience of the
world and frame our knowledge of it. However, in contrast to some of the more
extreme formulations of relativism and conventionalism, it is recognized that models are always models of something, where in general the existence of the “something” does not depend
29
on the models.
An interpretation of such a pragmatic realist view of measurement is illustrated
in Fig. 4.7. According to the account implicit in this figure, the existence of natural
reality is not denied, but neither is the fact that our various substantive and methodological theories and pragmatic concerns cause us to organize and prioritize experience in a particular way. This position is consistent with the Kantian metaphor that,
when we look at the world, we do so not directly but through a particular lens, or a
set of lenses, which affect what we see. Further, although it may never be possible
to look at reality without such lenses—and, as a corollary, it may never be possible
to draw a hard line of demarcation between theory and observation, or theoretical
and observational terms and concepts—we can still make considerable progress in
understanding the limitations of our own understanding by acknowledging and
examining our lenses as much as we are able, and where possible comparing them
to the lenses of others. Figure 4.7 visually illustrates the “lens” metaphor in the case
of measurement: when we attempt to use measurement processes to look at some
aspect of the world, what we see is jointly determined by the actual state of affairs
and how we have chosen, explicitly or implicitly, to model (at least) the general
property, the measurand, the environment, and the measurement process.
30
Thus, it is not necessary for properties to exist independently of consciousness
for them to be real and measurable. This point is perhaps most easily made with
respect to psychosocial properties such as reading comprehension ability or wellbeing. In these cases, the definitions of the properties are indexed to particular sets
of sociohistorical conditions, and thus their definitions are at least in part delineated
by contextually and pragmatically driven frames of reference. Additionally, in the
case of well-being (and other psychosocial properties such as desires, motivations,
attitudes, and physical experiences such as pain and hunger), the existence of the
29 Sometimes the very existence of modeled phenomena actually does depend to at least some
extent on models, which then become illocutionary (Austin, 1975). For example, whether a set of
neurophysiological facts about an individual make it more difficult for that individual to focus
attention over long periods of time compared to other individuals is arguably a model-independent
fact about that individual, but whether the individual has attention-deficit hyperactivity disorder
(ADHD) is a fact about how that individual has been labeled by other individuals, and is therefore
at least partially dependent on a model of ADHD.
30 Figure 4.7 is intended only as a rough visual representation of the role of models in producing
measurement results, not as a representation of the whole measurement process, which (of course)
involves actually performing measurements, which entails more than simply looking at reality
through the lens of models. Also, the concepts used in the figure—, etc.—will be explained further in later chapters.
4.5 A preliminary synthesis: model-dependent realism
mately caused by or supervene upon low-level cognitive and neurophysiological
processes.
This realism does not necessarily entail a naïve correspondence view of truth that
neglects the role of models in knowledge, nor does it deny that our conceptual
schemes, models, and linguistic frameworks actively shape our experience of the
world and frame our knowledge of it. However, in contrast to some of the more
extreme formulations of relativism and conventionalism, it is recognized that models are always models of something, where in general the existence of the “something” does not depend
29
on the models.
An interpretation of such a pragmatic realist view of measurement is illustrated
in Fig. 4.7. According to the account implicit in this figure, the existence of natural
reality is not denied, but neither is the fact that our various substantive and methodological theories and pragmatic concerns cause us to organize and prioritize experience in a particular way. This position is consistent with the Kantian metaphor that,
when we look at the world, we do so not directly but through a particular lens, or a
set of lenses, which affect what we see. Further, although it may never be possible
to look at reality without such lenses—and, as a corollary, it may never be possible
to draw a hard line of demarcation between theory and observation, or theoretical
and observational terms and concepts—we can still make considerable progress in
understanding the limitations of our own understanding by acknowledging and
examining our lenses as much as we are able, and where possible comparing them
to the lenses of others. Figure 4.7 visually illustrates the “lens” metaphor in the case
of measurement: when we attempt to use measurement processes to look at some
aspect of the world, what we see is jointly determined by the actual state of affairs
and how we have chosen, explicitly or implicitly, to model (at least) the general
property, the measurand, the environment, and the measurement process.
30
Thus, it is not necessary for properties to exist independently of consciousness
for them to be real and measurable. This point is perhaps most easily made with
respect to psychosocial properties such as reading comprehension ability or wellbeing. In these cases, the definitions of the properties are indexed to particular sets
of sociohistorical conditions, and thus their definitions are at least in part delineated
by contextually and pragmatically driven frames of reference. Additionally, in the
case of well-being (and other psychosocial properties such as desires, motivations,
attitudes, and physical experiences such as pain and hunger), the existence of the
29 Sometimes the very existence of modeled phenomena actually does depend to at least some
extent on models, which then become illocutionary (Austin, 1975). For example, whether a set of
neurophysiological facts about an individual make it more difficult for that individual to focus
attention over long periods of time compared to other individuals is arguably a model-independent
fact about that individual, but whether the individual has attention-deficit hyperactivity disorder
(ADHD) is a fact about how that individual has been labeled by other individuals, and is therefore
at least partially dependent on a model of ADHD.
30 Figure 4.7 is intended only as a rough visual representation of the role of models in producing
measurement results, not as a representation of the whole measurement process, which (of course)
involves actually performing measurements, which entails more than simply looking at reality
through the lens of models. Also, the concepts used in the figure—
4.5 A preliminary synthesis: model-dependent realism