243
person, which effect I call its interpretant, that the latter is thereby mediately determined by the former”, Albert Atkin (2013) is explicit that “we can think of the sign
as the signifier, for example, a written word, an utterance, smoke as a sign for fire
etc. The object, on the other hand, is best thought of as whatever is signified, for
example, the object to which the written or uttered word attaches, or the fire signified by the smoke”. Hence signs, and data, can be entities of a language—like
English words—but can also be nonlinguistic entities—like smoke—which are distinguishable (smoke vs. non-smoke: the syntactic layer), are possibly somehow
connected to something else (the fire: the semantic layer), and are possibly relevant
for decision-making (call firefighters: the pragmatic layer).
5
Analogously, data
acquired by means of measurement is not linguistic; rather, it is represented by linguistic entities—typically, numerals and names of units in the case of quantitative
properties—but it is not linguistic as such, as was shown in the discussion of values
of properties in Chap. 6.
8.1.2 A semiotic perspective on measurement
An analysis of measurement as a process of (syntactic, semantic, and possibly pragmatic) information acquisition throws further light on the connections between the
empirical and informational components of the process, which we now interpret
according to the model-based realism introduced in Sect. 4.5. In turn, this will allow
us to finally identify what fundamentally characterizes measurement itself.
Let us consider the model of direct measurement presented in Sect. 7.3 from the
perspective of the distinctions between and relations among syntax, semantics, and
pragmatics. Again, examples could be the measurement of temperature by means of
an alcohol thermometer, which operates by transducing temperatures to positions of
the upper surface of the alcohol in the instrument’s tube, or the measurement of
reading comprehension ability by means of a test, which operates by transducing
abilities to patterns of responses.
The core empirical component of a measuring instrument is the transducer that
interacts with the effective property and produces a transduced property, i.e., an
instrument indication, in response. With the assumption that the transduced property is observable or can be made observable, whether by human beings or technological devices, this input-output behavior is the bottom line of the process: if
5 Since data can be nonlinguistic, a further layer may be introduced for distinguishing between data
and its representation. Depending on the context, smoke might be represented by the English word
“smoke”, the Italian word “fumo”, etc. The fact that data, if it is not a linguistic entity, needs to be
represented in order to be manipulated, communicated, etc. has nothing to do with the relations
among syntactic, semantic, and pragmatic information. In what follows we do not deal with the
problem of representation of data (an example is about the decimal representation of non-integer
numbers: should the number 3/2 be represented as “1.5” or “1,5”?). The reader interested in notational issues may refer to the SI Brochure, Section 5, “Writing unit symbols and names, and
expressing the values of quantities” (BIPM, 2019).
8.1 Introduction
person, which effect I call its interpretant, that the latter is thereby mediately determined by the former”, Albert Atkin (2013) is explicit that “we can think of the sign
as the signifier, for example, a written word, an utterance, smoke as a sign for fire
etc. The object, on the other hand, is best thought of as whatever is signified, for
example, the object to which the written or uttered word attaches, or the fire signified by the smoke”. Hence signs, and data, can be entities of a language—like
English words—but can also be nonlinguistic entities—like smoke—which are distinguishable (smoke vs. non-smoke: the syntactic layer), are possibly somehow
connected to something else (the fire: the semantic layer), and are possibly relevant
for decision-making (call firefighters: the pragmatic layer).
5
Analogously, data
acquired by means of measurement is not linguistic; rather, it is represented by linguistic entities—typically, numerals and names of units in the case of quantitative
properties—but it is not linguistic as such, as was shown in the discussion of values
of properties in Chap. 6.
8.1.2 A semiotic perspective on measurement
An analysis of measurement as a process of (syntactic, semantic, and possibly pragmatic) information acquisition throws further light on the connections between the
empirical and informational components of the process, which we now interpret
according to the model-based realism introduced in Sect. 4.5. In turn, this will allow
us to finally identify what fundamentally characterizes measurement itself.
Let us consider the model of direct measurement presented in Sect. 7.3 from the
perspective of the distinctions between and relations among syntax, semantics, and
pragmatics. Again, examples could be the measurement of temperature by means of
an alcohol thermometer, which operates by transducing temperatures to positions of
the upper surface of the alcohol in the instrument’s tube, or the measurement of
reading comprehension ability by means of a test, which operates by transducing
abilities to patterns of responses.
The core empirical component of a measuring instrument is the transducer that
interacts with the effective property and produces a transduced property, i.e., an
instrument indication, in response. With the assumption that the transduced property is observable or can be made observable, whether by human beings or technological devices, this input-output behavior is the bottom line of the process: if
5 Since data can be nonlinguistic, a further layer may be introduced for distinguishing between data
and its representation. Depending on the context, smoke might be represented by the English word
“smoke”, the Italian word “fumo”, etc. The fact that data, if it is not a linguistic entity, needs to be
represented in order to be manipulated, communicated, etc. has nothing to do with the relations
among syntactic, semantic, and pragmatic information. In what follows we do not deal with the
problem of representation of data (an example is about the decimal representation of non-integer
numbers: should the number 3/2 be represented as “1.5” or “1,5”?). The reader interested in notational issues may refer to the SI Brochure, Section 5, “Writing unit symbols and names, and
expressing the values of quantities” (BIPM, 2019).
8.1 Introduction