Physics and Reality PDF
Physics and Reality PDF
Physics and Reality PDF
1. Introduction
196
M. BUNGE
PHYSICS A N D REALITY
197
198
M. BUNGE
199
200
M. BUNGE
Mediate
reference
v
I
Representation
J
Thus, the mediate referent of classical thermodynamics (thermostatics) is any reasonably insulated physical system, which is
represented as a fluid in a perfectly closed container (the fluidcum-walls being the theorys model). Of course there are no such
systems in nature, with the exception of the universe as a whole.
201
But any system contained in a calorimeter satisfying approximately the closure condition can be said t o be a mediate referent of
thermostatics.
In plain words : physics intends t o represent reality but i t does
it in a hypothetical, roundabout, and partial way. (In more
detail : a physical theory T is about a definite conceptual class Uthe universe of discourse of T. U corresponds to-but is notsome subset Z of reality. The phrase (( T refers immediately to U R
means that the formulas of T hold, by stipulation, for any element
of U, i.e. for the model. And the expression (( T refers mediately
to C Dmeans t h a t U is assumed to correspond to Z C R , whether
or not the formulas of T remain true when the members of U
occurring in them are replaced by the corresponding members of Z.
If the theory not only refers t o facts but in addition does it in
a verisimilar way, so much the better. In this case, i.e. for a
good theory, we have
where ((
=H
&
202
M. BUNGE
203
initially free object are altered. And the reason for regarding
objective meanings as more fundamental than operational ones
is this: anything t h a t is made u p of a physical object interacting
with a piece of apparatus constitutes a third, more inclusive,
physical system entitled to be studied as such and consequently
calling for objectively meaningful symbols. Only the test of theoretical statements concerning this wider system will require the
interpretation of some of the terms occurring in them in terms
of laboratory operations.
What has just been said holds for quantum physics as well as
for classical physics. The chief difference is that in classical
physics the theories of measurement are advanced enough to
enable us to calculate (predict) the disturbances introduced by
specified empirical operations, whereas no such mature theories
exist for the atomic and subatomic domains. In other words, in
classical physics we can account for the objective difference between a natural system and another object of the same kind interacting with our physical means of observation : the interaction is
incorporated into the law statements and the result of the calculation can be checked empirically for the system under measurement. If the prediction calculated by means of the theory of
measurement is borne out by measurement, the theory concerning
the natural object is regarded as confirmed (not as verified). On
the other hand, the quantum-mechanical theory of measurement is
not yet capable of yielding similar results. (We shall here disregard the mutually incompatible claims that the available theory
does its full job and that no such theory is conceivable because the
object-subject interaction is ultimately unanalyzable, i.e. irrational.) In any case, there are physical and epistemological
diRerences between a natural system and a measured one, and
correspondingly between the objective and the operational interpretation of a physical symbol.
Our semantical distinction is meaningless in the context of the
operationalist philosophy of physics. So much the worse for this
philosophy, since that distinction is in fact employed in physics,
notwithstanding the strenuous efforts made to reduce every idea
to perceptions and operations taking place in a conceptual vacuum.
204
M. BUNGE
PHYSICS A N D REALITY
205
in the hamiltonian : the latter will contain just the position, time,
and momentum coordinates of the supposedly autonomously
existing system-or rather of a sketch of it, such as an oscillator.)
Yet many physicists, misled by what used to be a fashionable
philosophy, smuggle into the theorems something that was missing
in the initial assumptions-namely,
a measurement device and
eventually even its operator, with his thoughts and his unpredictable intentions. This is how Heisenbergs relations are frequently interpreted, although no symbol representing measurement
operations-let alone mental events-occurs in the axioms from
which they are derived. Another example: the possible energy
states of a free helium atom are, in the same vein, interpreted as
the possible results of energy measurements-which measurements
would involve perturbations that had not been assumed to begin
with, i.e. when writing down the Schrodinger equation for the
helium atom. This equation contains no variable referring to the
structure and behavior of the phony measurement device, and i t
is only to comply with a philosophy accepted in advance that the
theorems are interpreted in a way unwarranted by the initial
assumptions. In brief, the solution t o the original problem is
somehow interpreted as the solution to an entirely different problem-a shift designed to introduce the Operator into the innermost recesses of nature. Our quantum philosopher-physicist is
thereby making use of a privilege formerly reserved to theologians :
namely, that of ((concluding)) from one statement to a second
statement referring to an altogether disjoint universe of discourse.
Let us take a closer look at this commonly practised yet ill-studied
strategy.
5. Conceptual Unity-And
206
M. BUNGE
meaning shifts characterize the usual (phenomenalistic, operationalistic, idealistic) interpretations of both the quantum theory
of ((particles o and the quantum field theory, which are thereby
rendered what shall be called semantically inconsistent. Since the
trivial yet important concept of semantical consistency, or conceptual unity, does not seem to have been analyzed, a brief excursus
will be convenient a t this point. A fuller treatment is given elsewhere l.
I t is a desideratum for every theory to possess both formal and
semantical unity. The former consists in the systems logical
togetherness, i.e. in being a hypothetico-deductive system rather
than an arbitrary heap of formulas. The semantical consistency
or conceptual unity of a factual theory boils down to this: the
system must be about some (nonempty) class which, far from
being an arbitrary collection, is characterized by certain mutually
related properties. Let us give a more precise characterization of
semantical consistency.
To begin with, the conceptual unity of a theory requires a
common reference of its formulas to some collection of objects.
In the case of a physical theory, this collection is not an arbitrary
set but a natural (nonarbitrary) class of physical objects. The
class of objects the theory refers to is the latters universe of
discourse. Thus, the universe of discoursz of fluid mechanics is
the class of all fluids: the theory assigns the latter certain properties every one of which it represents by a certain predicate.
It is taken for granted that the universe of discourse or reference
set is not void, and it is hypothesized that the members of it can
be paired to external objects in such a way that the theory holds
a t least approximately. Such a reference to external objects can
be indirect and even false (see Sec. 3), but some reference t o physical objects is always presupposed in a physical theory, and this
is why it is called physical rather than, say, psychological. Theories carry the names of their ultimate referents, even if the latter
turn out not to exist; thus, a theory referring to hyperons will
be called a hyperon theory. On the other hand, a (meta)statement such as ((The statements of quantum mechanics are not
Cf. S R , Sec. 7.2.
PHYSICS A N D REALITY
207
208
M. BUNGE
PHYSICS A N D REALITY
209
210
M. BUNGE
but it does work for the physical universe, and every retreat from
physicalism in the realm of physics is a return to prescientific
anthropocentrism. Why misplace the human mind: is it not a
system of functions of certain bodies made up of atoms, and is it
not enough to credit the human mind with the invention of theories,
the planning of tests, and the interpretation of the latters outcomes ?
(We should note in passing t h a t a realistic interpretation of
the quantum theory does not require renouncing its present fundamental stochastic character. In other words, one does not need
t o introduce extra hidden variables in order to restore objectivity
in t h e quantum domain: t h e hidden variables are already there.
Only, they are miscalled obseruables, although no one can seriously
claim that any of the fundamental variables of the quantum
theories are strictly, i.e. directly observable or measurable.
Hidden variables, in the sense of nonstochastic-nonfluctuating,
scatter-less-magnitudes, are sufficient, yet not necessary, to produce a nonstochastic theory resembling classical dynamics. But
such neoclassical concepts are very likely insufficient, and certainly dispensable, to build the much-needed semantically consistent and thoroughly physical interpretation(s) of the quantummechanical formalism. One should not mix the problems of
reality and objectivity with the problem of determinism : a
realist can consistently maintain an indeterministic stand to any
extent, just as a subjectivist can be as much of a determinist as he
pleases. For realism the precise behavior of physical objects is
irrelevant as long as they can walk alone.)
In order to restore realism in physics all we need is to reinterpret the present formalisms of the quantum theory abiding by the
rules of semantical consistency and keeping in mind the goal of
producing a physical rather than a psychological theory of the
microphysical world. This is possible now, without modifying the
present formalisms-which
need repairs for different purposes.
Such a realistic interpretation of the available frameworks is
Cf. the authors Causality: The Place of the Causal Principle in
Modern Science, 2nd ed. (Cleveland and New York : Meridian Books, 1963),
Appendix.
211
212
M. BUNGE
t o it. Thus, atomic theories are not about spectroscopic observations although they participate (alongside other theories) in the
explanation of such data.
Th co r y
Modcl
l,,,d;,
rcfercncc I
rcfercncc
I
1\
i
\
'
Obscrvcd
\
0
I
/
Obscrvahle
PHYSICS A N D REALITY
213
optics, which play here an instrumental rather than an explanatory or substantive role. Another example : any theory of (( elementary o (( particles )) refers immediately to certain unobservables
suspected (hypothesized) t o be entities (real existents), but it only
supplies a hypothetical and most likely a rather crude model of
them. And the evidence relevant t o such a theory-e.g., a set of
tracks on a nuclear photographic plate-differs in nature from the
referents of the theory: the tracks are not referred t o a t all by
the theory and these data become an evidence relevant t o the theory
on condition that they be interpreted in the light of another body
of theory (notably classical mechanics and some theory concerning
the passage of electrically charged particles through matter).
In this respect the job of the physicist is not unlike the task of
the paleontologist, the historian, or even the detective: in all
these cases unseen facts are hypothesized and such hypotheses
and systems of hypotheses are tested through the observable
traces left by the presumptive criminal (extinct animal, hero, or
proton), which traces become evidences only in the light of instrumental or auxiliary hypotheses and/or theories concerning the
possible mechanisms whereby the traces might have been produced ; clearly, the theory under test may take place in such an
explanation, i.e. it may contribute t o produce its own evidence.
(So far we have purposefully avoided talking of phenomena as
data, much less as evidence relevant t o physical theories. The
reason is this. What philosophers call a phenomenon is an event
occurring in connection with some cognitive subject : phenomena
are what appear to us, humans, whence no humans, no phenomena.
Phenomenalism is the doctrine according t o which the world is the
set of appearances ; in particular, physical reality would be the set
of observations conducted by physicists. The programme of
phenomenalism, shared t o a large extent by operationalism, is the
construction of physical objects as systems of appearances. This
programme has failed and is unfeasible. There are various reasons
for rejecting phenomenalism, among others the following. Firstly,
physics is not interested in what appears to me, or in what seems
to me t o be the case : physics is an attempt t o transcend subjectivity,
t o go beyond perspectivism. Secondly, most phenomena involve
214
M. BUNGE
215
Prediction
Datum 1
Theory 2
(e.g.,chargevalue) (instrumental)
+---- ---Comparison---
Datum 2
(e.g.,trackcurvature)
-----
To sum up : (i) fundamental physical theories have no observational content, i.e. they contain no purely observation statements,
and consequently they cannot be reduced to sets of data or even
to data-processing devices ; (ii) there is no theory-free evidence in
physics. If this much is accepted, we need not confuse reference
with evidence.
7 . Rules of Interpretation
If the preceding analysis is accepted, then it must be acknowledged that in physics we meet more than one kind of rule of
interpretation (semantical rule). The mathematical formulas of
physics can be read with the assistance of interpretation rules of
two kinds : referential and evidential. A referential interpretation
rule (RIR) establishes a correspondence between some of the
theorys nonformal symbols, and its referent. Consequently, a
rule of this kind contributes to the (core) meaning of the theory ;
in the ideal case of a very simple theory, the set of its referential
interpretation rules makes up the full physical meaning of the
theory. On the other hand a n evidential interpretation rule (EIR)
links a low level theoretical term to some observable entity or
trait, such a visible clock.
Cf. S R . Sec. 8.4.
216
M. BUNGE
ttical symbol
/ 9 E IR
RIR(f
Real referent
Experience
PHYSICS A N D REALITY
217
218
M. BUNGE
Expcriciicc
PfIYS I C S
Actual facts
8. Concluding Remarks
Every theory is about objects of some kind-which it identifies
as members of a universe of discourse U-to which it assigns
certain definite and basic properties Pi constituting the predicate
219
220
M. BUNGE
PHYSICS A N D REALITY
221
shades of subjectivism are left behind as so many partial viewssimplistic views-of our science. Note that these views are not
here discarded on the strength of philosophical tenets and arguments of a traditional kind but with the help of mathematical logic,
semantics and methodology, the very tools which were once
thought to support the subjectivistic philosophies of physics.
The vacuum left by the (logical) passing away of the abovementioned philosophies of physics should be filled by building a
theory of knowledge subsuming and expanding the seeds of truth
contained in the former doctrines, as well as the realistic hypotheses presupposed and suggested by science 2. Such a critical
or scientific realism could assist (and in turn be tested by) the
construction of a thoroughly physical (rather than psychological)
interpretation of the mathematical formalisms of the quantum
theory. The new realistic epistemologist should both stick his
neck out and be prepared t o have it cut : he should advance daring
(yet grounded) assumptions, none of which he should regard as
uncontroversial. Thus, while the naive realist may take the reality
of electrons for granted, the critical realist will say t h a t present-day
physics assumes t h a t electrons are real things4.e. he hypothesizes
that the electron concept has a concrete counterpart but, a t the
same time, he would not be too surprised if this assumption turned
out t o be false and electrons were replaced by something else.
Needless to say, whatever form the new critical realist epistemology may take, it would fail t o meet the standards of scientific
research and consequently i t would fail t o help this enterprise if it
were conceived as one more ism, i.e. as a set of tenets beyond
criticism and above science. Wanted: a name for this nascent
epistemology, one not ending in ism, for whatever ends in ism is
apt t o put a n end t o the quest for truth.
Cf. S R , Sec. 5.9.
See the following recent defenses of critical realism : P. BERNAYS
contribution to the discussion on quantum mechanics in the Revue de
(1 Matter Still Largely
Mitarphysique et de Morale, avril-juin 1062 ; H. FEIGL,
Conjectures
Material )), Philosophy of Science, 29, 39 (1962) ; K . K. POPPER,
and Refutations (London : Routledge and Kegan Paul ; New York : Basic
Books, 1963) ; and J.J.C. SMART,
Philosophy and Scientific Realism (London :
Routledge and Kegan Paul, 1963).
a
222
M. BUNGE
Abstract
A semantical and methodological analysis of physical theories is performed in order to And out their relation to reality and to human experience.
I t is shown that every physical theory refers immediately t o an idealized
model of what is supposed to be a piece of reality-the mediate referent
of the theory. Two kinds of physical interpretation of physical symbols
are distinguished : objective and operational, and the difference between
reference and evidence is stressed. I t is claimed that for a theory to be
physically meaningful i t is necessary that it includes rules of objective
reference. In some cases a theory contains, in addition, evidential reference
rules, i.e. correspondences between some of its concepts and empirical
items. But the test of any theory requires, rather than operational rulcs,
further theories t h a t can bridge the gap between what the theory refers t o
(e.g., the passage of a cosmic ray) and its remote symptom (e.g., the click
of a counter). The conditions for the semantical consistency are sketched
and i t is argued that there does not exist a semantically consistent interpretation of any of the formalisms of quantum theory. From this analysis
arguments in favor of critical realism are drawn.