254
8.3.4 … and a focus on producing explicitly justifiable
information
Science aims at the development of knowledge, where, following Plato, knowledge
is commonly understood as justified true belief.
9
But although science aims at truth,
it cannot guarantee it, as plainly illustrated by its history, and thus the truth or falsity
(and more generally the quality) of any given scientific theory cannot be a definitional component of what makes it scientific: a theory can be of low quality, and
even eventually admitted to be false, and nevertheless can be scientific. Although
discussions of the definition and essential features of science are still ongoing, it is
generally agreed that scientific theories must be explicitly justifiable, in that their
logical and evidentiary grounding must be clear and publicly evaluable (see, e.g.,
Hansson, 2017). Thus, science is characterized by its structure rather than its outcomes; it is no contradiction to say that a theory is both scientific and false, but it
would be a contradiction to say that a theory is both scientific and untestable, even
if the theory were true.
10
While approaching the end of the path we have followed in this book, we state
our belief that this feature of science applies equally well to measurement, and is in
fact the sufficient condition we were seeking, for complementing the necessary conditions introduced in Chap. 2: the trustworthiness of measurement results is not
earned solely by their objectivity and intersubjectivity, but first of all by their justification. More explicitly, the result of the evaluation of a property can be claimed to
be a measurement result only on the condition that in principle it is possible to
explain how it was obtained with sufficient clarity to allow its critical analysis by all
9 “According to this account, the three conditions—truth, belief, and justification—are individually
necessary and jointly sufficient for knowledge of facts” (Steup & Ram, 2020: 2.3).
10 A clear and simple example of this fundamental characterization of science is given, by difference, by Daniel Dennett: “There are many strategies, some good, some bad. Here is a strategy, for
instance, for predicting the future behavior of a person: determine the date and hour of the person’s
birth and then feed this modest datum into one or another astrological algorithm for generating
predictions of the person’s prospects. This strategy is deplorably popular. Its popularity is deplorable only because we have such good reasons for believing that it does not work. When astrological
predictions come true this is sheer luck, or the result of such vagueness or ambiguity in the prophecy that almost any eventuality can be construed to confirm it. But suppose the astrological strategy
did in fact work well on some people. We could call those people astrological systems—systems
whose behavior was, as a matter of fact, predictable by the astrological strategy. If there were such
people, such astrological systems, we would be more interested than most of us in fact are in how
the astrological strategy works—that is, we would be interested in the rules, principles, or methods
of astrology. We could find out how the strategy works by asking astrologers, reading their books,
and observing them in action. But we would also be curious about why it worked. We might find
that astrologers had no useful opinions about this latter question—they either had no theory of why
it worked or their theories were pure hokum. Having a good strategy is one thing; knowing why it
works is another” (1987: p.16). We claim exactly the same of measurement: that its results work
(in some sense) is not enough; we want to know why they work. And this requires “opening the
box” of the process and examining its structure and functioning.
8 Conclusion
8.3.4 … and a focus on producing explicitly justifiable
information
Science aims at the development of knowledge, where, following Plato, knowledge
is commonly understood as justified true belief.
9
But although science aims at truth,
it cannot guarantee it, as plainly illustrated by its history, and thus the truth or falsity
(and more generally the quality) of any given scientific theory cannot be a definitional component of what makes it scientific: a theory can be of low quality, and
even eventually admitted to be false, and nevertheless can be scientific. Although
discussions of the definition and essential features of science are still ongoing, it is
generally agreed that scientific theories must be explicitly justifiable, in that their
logical and evidentiary grounding must be clear and publicly evaluable (see, e.g.,
Hansson, 2017). Thus, science is characterized by its structure rather than its outcomes; it is no contradiction to say that a theory is both scientific and false, but it
would be a contradiction to say that a theory is both scientific and untestable, even
if the theory were true.
10
While approaching the end of the path we have followed in this book, we state
our belief that this feature of science applies equally well to measurement, and is in
fact the sufficient condition we were seeking, for complementing the necessary conditions introduced in Chap. 2: the trustworthiness of measurement results is not
earned solely by their objectivity and intersubjectivity, but first of all by their justification. More explicitly, the result of the evaluation of a property can be claimed to
be a measurement result only on the condition that in principle it is possible to
explain how it was obtained with sufficient clarity to allow its critical analysis by all
9 “According to this account, the three conditions—truth, belief, and justification—are individually
necessary and jointly sufficient for knowledge of facts” (Steup & Ram, 2020: 2.3).
10 A clear and simple example of this fundamental characterization of science is given, by difference, by Daniel Dennett: “There are many strategies, some good, some bad. Here is a strategy, for
instance, for predicting the future behavior of a person: determine the date and hour of the person’s
birth and then feed this modest datum into one or another astrological algorithm for generating
predictions of the person’s prospects. This strategy is deplorably popular. Its popularity is deplorable only because we have such good reasons for believing that it does not work. When astrological
predictions come true this is sheer luck, or the result of such vagueness or ambiguity in the prophecy that almost any eventuality can be construed to confirm it. But suppose the astrological strategy
did in fact work well on some people. We could call those people astrological systems—systems
whose behavior was, as a matter of fact, predictable by the astrological strategy. If there were such
people, such astrological systems, we would be more interested than most of us in fact are in how
the astrological strategy works—that is, we would be interested in the rules, principles, or methods
of astrology. We could find out how the strategy works by asking astrologers, reading their books,
and observing them in action. But we would also be curious about why it worked. We might find
that astrologers had no useful opinions about this latter question—they either had no theory of why
it worked or their theories were pure hokum. Having a good strategy is one thing; knowing why it
works is another” (1987: p.16). We claim exactly the same of measurement: that its results work
(in some sense) is not enough; we want to know why they work. And this requires “opening the
box” of the process and examining its structure and functioning.
8 Conclusion