Beruflich Dokumente
Kultur Dokumente
STUDIES IN EPISTEMOLOGY,
Managing Editor:
Editors:
VOLUME 309
A COLLECTION
OF POLISH WORKS
ON PHILOSOPHICAL
PROBLEMS OF .TIME
AND SPACETIME
Edited by
HELENA EILSTEIN
The Institute 0/ Philosophy and Sociology,
Polish Academy 0/ Science, Warsaw, Poland
HELENA EILSTEIN
PREFACE vii
JERZYGOLOSZ
MOTION, SPACE, TIME 1
LESZEK M. SOKOLOWSKI
QUANTUM SPACETIME AND THE PROBLEM
OF TIME IN QUANTUM GRAVITY 23
MICHAL HEUER
TIME AND PHYSICS - A NONCOMMUTATIVE
REVOLUTION 47
JAN CZERNIA WSKI
FLOW OF TIME AS ASELECTION RULE IN GENERAL
RELATIVITY 57
TOMASZ PLACEK
BRANCHING FOR A TRANSIENT TIME 73
HELENA EILSTEIN
AGAINST DETENSERS (NOT FOR TENSERS) 93
STEFAN SNIHUR
ON EXISTENCE OF THE FUTURE 127
ANDRZEJ P6LTA WSKI
THE PROBLEM OF TIME IN THE PHILOSOPHY
OF ROMAN INGARDEN 137
HELENA EILSTEIN
PREFACE
Abstract. The paper discusses the properties of spacetime we study by analyzing the phenomenon of
motion. Of special interest are the spacetime symmetries. the spacetime structures and the ontological
status of spacetime. These problems are considered on the grounds of the c1assical theories of motion
contained in Newtonian physics, special and general theory of relativity. The controversy between an
absolute and a relational conception of motion and its ontological implications are also analyzed.
1. INTRODUCTION
Because space and time are not directly accessible to our senses, we are forced to
study them indirectly through phenomena taking place in them. Such justification is
needed by substantivalists, who admit that space and time exist independently of
material world, but is not needed by relationists and advocates of property view·,
who deny that space and time are substances. The necessity of resorting to physical
phenomena is for them a natural consequence of accepted ontological assumptions.
Motion is one of the most interesting phenomena, which can provide us with
information of space and time. Searching for an adequate theory of motion helps us
to understand space and time: their properties, structures they are endowed with and
relations between them. In this paper I would like to analyze this problem firstly in
the nonrelativistic theory and then in the relativistic theory. Last of all I would like
to discuss ,he controversy between the absolute and the relational conception of
motion and its ontological consequences.
There is, however, one problem, that will not be considered in the paper. This is
the problem of time reversibility of physical phenomena. All known theories of
motion are time reversible, but the problem of time reversibility of physical
phenomena cannot be discussed on the sole ground of the analysis of the
phenomenon of motion.
If we want to describe a motion of bodies, we must decide what this motion is
related to and what properties it has. The latter question concerns the spacetime
symmetries of the intended theory of motion, the former - the problem whether we
want to describe the motion of bodies with respect to space and time (if necessary
- spacetime) or to other bodies. Each of these choices assurnes some properties of
space and time and the test of adequacy of the obtained theory of motion teils us
whether our assumptions are right or not (by adequacy of a theory I understand its
ability to explain and predict physical phenomena).
Let's consider alternative ways of building theories of motion. 1 will begin with the
relational and the absolute conceptions of motion. The relational conception
of motion can be expressed in the following way:
REL Each motion of bodies is relative to other bodies or takes place relative
to adefinite structure which is determined by the distribution of mass in the
Universe.
ABS Each adequate theory of motion should contain in its equations at least
one of the absolute (that is, relating to space or spacetime, and not to other bodies)
quantities, such as location, velocity, acceleration, etc.
into a wide vesseI beneath it. With the ship standing still, observe carefully how the
littIe animals fly with equal speed to all sides of the cabin. The fish swim indifferently
in all directions; the drops fall into the vessel beneath; and, in throwing something to
your friend, you need throw it no more strongly in one direction than another, the
distances being equal; jumping with your feet together, you pass equal spaces in every
direction. When you have observed a11 these things carefully (though there is no doubt
that when the ship is standing still everything must happen in this way), have the ship
proceed with any speed you like, so long as the motion is uniform and not fluctuating
this way and that. You will discover not the least change in a11 the effects named, nor
could you tell from any of them whether the ship was moving or standing still. (Galileo
1632,pp.186-187)
The result of this discovery was an important physical principle called the
principle of Galilean relativity which in its classical formulation says, that
mechanical phenomena do not distinguish any of inertial frames, rectilinearly and
uniformly moving relative to one another. This principle together with the
requirement of absoluteness of time has led to the Galilean transformations, which
correspond to the symrnetry group of Newtonian dynarnics:
(GAL)
t -Y t' = t + constant
(1)
Thus, the acceleration appearing in the second law of dynamics is the absolute
acceleration (acceleration relative to space) and Newtonian dynamics is an absolute
theory of motion. This fact has not been noticed by Newton's opponents and some
of their commentators;4 Berkeley and Mach, criticizing Newton's absolute space,
did not propose any alternative theory, which could link up the inertial structure with
the distribution of mass in the Uni verse. The problem of the ontological
consequences of the absoluteness of motion will be discussed in § 4, whereas now
I would like to analyze exactIy the Galilean spacetime, introduced by (GAL).
Traditionally, it was assumed that spacetime symmetries of a theory are
represented by the symmetries of its equations. The symmetry mappings of the
Newton's second law (1), for example, assurne the form (GAL). At present
we know, however, that Newtonian mechanics, like many others physical theories,
can be expressed in a generally covariant form and thus we cannot identify
symmetries of theory's equations with the symmetry of that theory5. E.g. Newton's
second law assurnes the following generally covariant form:
(2)
where r ~k are coefficients of a flat affine connection, that is, of a connection for
which there exists aglobai coordinate system, in which r~k =0 (i,j,k = },2,3,4). The
coordinate systems satisfying this condition are just inertial frames. Equations of the
form (2) do not change under any differentiable transformation.
To introduce the concept of spacetime symmetry of a certain theory we must
distinguish between absolute and dynamical objects of that theory.
The absolute objects Ai are those that are not affected by the interactions described
in the theory. They characterize the fixed spacetime structure assumed in the theory
in question and are invariant with respect to the corresponding transformations.
The dynamical objects Pi characterize the physical content of its spacetime and can
be affected by the interactions described in the theory. Examples of absolute objects
are space metric and absolute time in the case of Newtonian mechanics, and the
metric of special theory of relativity (hereafter STR). The metric of GTR, affected
by the energy-momentum tensor, and the electromagnetic field tensor, affected by
the current density four-vector, are examples of dynamical objects. Models of any
physical theory T may be expressed in the following form:
metric hij for the three-dimensional instantaneous spaces. The principle of Galilean
relativity can now be expressed in the following form: the symmetry group
of Newtonian mechanics (M, r~k' tj , hij ) is the Galilean group (GAL).
The symmetries we are discussing inform us about important properties of space and
time in Newtonian physics. We have the following properties: the homogeneity of
space and time (expressed by the invariance of the absolute objects of Newtonian
mechanics under the spatial and temporal translations), the isotropy of space
(expressed by invariance of the absolute objects under the spatial rotation) and the
symmetry in respect of mirror image reflection. It is worth noting that, according to
Noether's theorem, every symmetry (in particular, every spacetime symmetry)
corresponds with a some conservation law. And so the invariance under temporal
translations corresponds with the energy conservation law, the invariance under
spatial translations implies the momentum conservation law, and the invariance
under the spatial rotations entails the angular momentum conservation law'.
The replacement of the equation (1) by the more general equation (2) does not
change absoluteness of Newtonian mechanics, for the affine connection appearing in
this last equation can be related in the Newtonian mechanics only to spacetime.
In the equation (2) we have also the absolute (relating to spacetime) acceleration
tfi/dr. The additional term r ~k (dx j /dt )( U /dt) appearing in this equation
describes the inertial forces acting in the noninertial reference frames. This term
vanishes in the inertial frames where r~k = o.
So Newtonian mechanics is the absolute theory of motion because the
acceleration appearing in its equations (1) (or (2)) relates to the inertial (or affine)
structure of spacetime. However, Newton understood this absoluteness in a different
way. He did not distinguish between the ontological absoluteness (the substantival
character) of space and the absoluteness in the sense of the existence of an absolute
(distinguished) reference frame. He thought that absoluteness of motion consists in
existence of an absolute (distinguished) reference frame:
Absolute motion is the translation of a body from one absolute place into another; and
relative motion, the translation from one relative place into another. Thus in a ship
under sail, the relative place of a body is that part of the ship which the body possesses;
or that part of the cavity which the body fills, and which therefore moves together with
the ship; and relative rest is the continuance of the body in the same part of the ship, or
of its cavity. But real, absolute rest, is the continuance of the body in the same part of
that immovable space, in which the ship itself, its cavity, and all that it contains, is
moved. (Newton 1729, p. 7)
It is surprising that Newton believed in the existence of such a frame and in that
absolute motion consist in the change of absolute position in this frame, although he
realized that he could not point it out:
And therefore as it is possible, that in the remote regions of the fixed stars, or perhaps
far beyond them, there may be some body absolutely in rest; but impossible to know,
from the position of bodies to one another in our regions, whether any of these do keep
the same position to that remote body; it follows that absolute rest cannot be determined
from the position of bodies in our regions. (Newton 1729, p. 8-9)
suppression of the time depending translations va. t. The symmetry mappings have
then the form:
(NEW)
t ~ t' = t +const
However, no physical laws suggest the existence of a distinguished reference
frame and the symmetries of Newtonian dynarnics are symmetries (GAL). Since the
concept of absolute position is not needed in the construction of adequate physical
theories, we can, in the spirit of Occam's razor, renounce the idea of existence of the
distinguished reference frame. Although the motion thus ceases to be absolute
in Newtonian sense, it remains absolute after the extension of the symmetries from
(NEW) to (GAL), since we have the absolute (nonrelational) acceleration
in the Galilean spacetime.
So Newton's critics were mistaken when they believed that it was enough to
reject the absolute (distinguished) reference frame for renouncing the absoluteness
of motion. It was possible to renounce the absoluteness in one way only - by
constructing a relation al theory of motion. Neither of the relationist constructed such
a theory.
It is worth noticing, that first relational theories of motion came into being only
in the second half of the 20th century. It happened so late probably because what
was needed for their creation was Hamiltonian formalism as weil as the awareness
that the domain of its application transcends Newtonian mechanics. AIthough these
theories do not represent any viable alternatives to Newtonian theory or to
Relativity, they are philosophically interesting. They were constructed by J. B.
Barbour and his co-workers.
A relationist looking for a nonrelativistic theory of motion has to choose between
two kinds of spacetime symmetries: wider ones with the set of invariants consisting
just of absolute simuItaneity and relative distances of particles, and the narrower
ones with the additional invariant in the form of a time interval. 9 The first ones are
called Machian symmetries, the second ones Leibnizian symmetries. 10
Machian symmetries have the form:
(MACH)
where R aß(t) is a time dependent orthogonal matrix and aa(t) andf(t) are arbitrary
smooth functions of time. The time parameter t in these theories has no metrical
significance, therefore any function l' = f (t) that does not change the order of events
in time (df / dt > 0) can stand for it. Time here is only a parameter used as a 'label'
for changing relative configurations of events. Time defined this way corresponds
to the idea of Leibniz and Mach according to which it is only a sequence of events.
MOTION, SPACE, TIME 7
(LEIB)
t ~ t' = t + const
Here time has a metrical significance because the time metric is an absolute
object of (LEIB). Symmetry mappings are restricted so as to preserve time interval.
It is worth pointing out two important issues. First of all, the introduced above
symmetries (MACH) and (LEIB) are merely symmetries of some possible theories
of motion and, sirnilarly, the absolute objects introduced above (the invariants of the
symmetries) are merely some special objects appearing in models of these theories.
So there is no need to introduce such entities as a Machian or Leibnizian spacetime,
because it may mean treating spacetime as a substance and is potentially misleading.
If anybody, in spite of all, wants to introduce such entities (Jike for example Earman
(1989b, pp. 27 - 31)), he should say that these spacetimes are not substances and his
text should not be interpreted literally.
My second comment concerns the interpretation of symmetry transformations
(MACH) and (GAL) allowed by relationists. A consistent relationist must interpret
these symmetry transformations passively as coordinate transformations. Indeed,
if he does not want to succumb to substantivalism, he cannot interpret them actively
as point transformations. For symmetry transformations, interpreted passively, only
mean that any physical system can be described equivalently in different coordinate
systems. This interpretation does not involve any ontological commitment
to spacetime. It is quite different in case of the active interpretation of symmetry
transformations. The latter implies, intuitively to speak, that it is possible (in the
light of the laws of physics) to relocate, reorient or boost physical systems in the
spacetime container. In other words, the standard interpretation of active
transformations is based on the assumption that spacetime points preserve their
identity though the material objects which are located in them change. Hence,
the active interpretation of transformations takes substantivalism for granted and
cannot be used by relationists and advocates of property view.
It may seem that the symmetry mappings (LEIB) are more interesting than
(MACH), because relative velocities and relative accelerations of bodies are
invariants of (LEIB) and thus the relationist may resort to them in his search for the
equations of motion. It turns out, however, that mappings (MACH) also have some
attractive properties, which induced Barbour to chose them as symmetries
for equations of motion. The point is that (MACH) allows for a free choice of j(t),
and this makes it possible to simplify some equations of the theory.
The idea of inventing a relational theory of motion expressed in the language
of relative distances between particles was presented by Barbour in 1974.
In the kinematical part of this conception the author introduces the relational
configuration space (RCS). If the Uni verse is assumed to consist of N point
particJes, the points of the RCS are all possible distinct relative configurations of
these particJes. Then, any continuous curve in the RCS forms a possible kinematical
history of the Uni verse and each point on any of these curves defines an 'instant of
8 JERZY GOLOSZ
time' in a given history. Time is thus defined by the history of the Universe as
a whole.
The dynamics is introduced to the ReS through an action principle for some
Lagrange function L. Barbour (1974) assurnes the Lagrange function of the form
(i,) =1, ... ,N):
L= 'P·F (3)
It is worth noticing that this equation has the Newtonian form with the
coefficient r= 11M'P, which can be interpreted, according to Barbour,
as a 'gravitational constant', determined by the actual distribution of matter in the
Universe. The second interesting result achieved by Barbour is that this model
'explains inertia (resistance of a body to rectilinear acceleration relative to
the remaining bodies in the Universe) solely in terms of relative distances and
relative velocities and demonstrates that a complete dynamics can be expressed in
such terms.' (Barbour 1974, p. 329). The weak point of this model, as pointed out by
Earman (1989b, p. 93), is that the restriction to one spatial dimension eliminates
rotation, the Achilles' tendon of relationism.
In the later article Barbour and Bertotti (1977) extended Barbour's analysis
to three spatial dimensions and modified the Lagrange function to reduce influence
of distant matter on the inertia of a given body.
Some predictions of Barbour's relational theory, like a non-Newtonian motion
of bodies in a Uni verse containing only a few particles, cannot be verified.
The theory of Barbour and Bertotti (1977) also predicts effects that at the time being
cannot be verified, as e.g. that the gravitational action of a spherical body is not the
same as if mass were concentrated at its center. It also predicts, contrary to GTR,
that the gravitational 'constant' Gis changing with time (G' IG - JO-JO per year).
MOTION, SPACE, TIME 9
These predictions, which can be verified, in some cases are in accordance with
experiments and in some cases are not. In contradistinction to Newtonian theory that
of Barbour and Bertotti gives correct values for the orbital period and the perihelion
advance of Mercury but it also predicts (1977, p. 21) mass-anisotropy effects which
are in crass contradiction with experiment.
One can agree with Earman (1989b, pp. 95 - 96) that more investigation
is needed before a final appraisal of theories of Barbour et al. is achieved.
Particularly, it would be necessary to formulate aversion of electromagnetism and
quantum mechanic with spacetime symmetries (MACH). The greatest merit
of the work of Barbour et al. consist in showing that interesting relational theories
are possible and in giving some insight into what such theories can be like.
Before Maxwell's electrodynamics it seemed that the principle of Galilean
relativity was in force for all physical phenomena. It turned out, however, that
Maxwell's equations are not invariant under transformations (GAL). At first it was
thus assumed that these equations single out adefinite reference frame, with respect
to which the velocity of light is c. That preferred reference frame was supposed to
be the one in which the hypothetical material medium, ether, was at rest.
Electromagnetic waves were interpreted as manifestations of vibrations of that
medium. Maxwell's equations were supposed to be satisfied in the ether frame and
to have a different form in other frames.
Since in each frame, which moves relative to the ether frame, the velocity
of light should have a value different than c it seemed that by comparing velocities
of light in different directions one can discover the relative motion of the Earth and
the ether. A relevant experiment was carried out by Michelson and Morley and gave
no result, what testified against the conception of ether. This experiment seemed
also to point out that the equations of electrodynamics have the same Maxwellian
form in all inertial frames. To save the existence of the ether, additional assumptions
were proposed. One of best known of them is the hypothesis which assurnes that
portions of the ether are dragged along by moving ponderable bodies. Another, even
better known hypothesis, submitted by H. Lorentz, states that bodies moving relative
to the ether undergo a contraction (in the direction of motion) and a time dilation.
The Lorentz's hypothesis enabled one to retain the conception of the absolute
(distinguished) reference frame but according to the null result of the Michelson-
Morley experiment this frame had to be experimentally undetectable.
However, new discoveries soon made it dear that the conception of ether
is redundant. First, Larmor and Poincare found out the symmetry mappings for
MaxweIl's equations. It turned out, that Lorentz's formulas for the length
contraction and time dilation resulted from Larmor and Poincare's mappings, but
because these mappings were radically different from the then presupposed (GAL)
spacetime symmetries, they were assumed to be only a formal property of the
Maxwell's equations. A situation changed in 1905 after Einstein proposed STR.
(POINC)
where ai , R ik = const, ltk g'ijR j, = gk/, gJ/ =-g22=-g33=-g44=1 and gij=O if i;t!j
(gij is the metric tensor).
Maxwellian electrodynamics was the first theory satisfying the .new, special
principle of relativity. Newtonian mechanics satisfied it only approximately,
for velocities which are small in comparison with the speed of light. In his first
papers on STR, Einstein proposed a new mechanics, invariant under (POINC).
The transformations (POINC) form a group and they introduce to spacetime
a four-dimensional geometry, which is called Minkowski geometry. Applying
the four-dimensional tensor calculus to STR, Minkowski presented a formalism
where the form of laws by itself guarantees their invariance under (POINC). This
calculus is the counterpart of the three-dimensional tensor calculus in Euclidean
space.
It is claimed sometimes that four-dimensional spacetime was introduced into
physics only with STR. As a matter of fact, however, a four-dimensional spacetime
can be introduced into Newtonian physics as well, but in it the hyperplanes
of simultaneous events are absolute (independent of any reference frame) and thus
the four-dimensional point of view is not necessary. In Minkowski spacetime space
and time cannot be separated this way. We must consider them to be united into one
entity - Jour-dimensional spacetime - and following Minkowski we have to give
up the view that space and time are independent.
The main part in the Minkowski geometry is played by the invariant LI .. of the
group (POINC), called the spacetime interval. It satisfies the equation
-2 ..
LI.=gijL1x'LIx'=c2
(t-to j -(x-xo)2 -(Y-Yo)2 -(z-Zo j 2 • .2
=CLJr-LJr.2
(5)
o (Lir = 0). They form the so-called light (or null) eone associated with the point 0
and have the property that either they can be reached by light signals sent from 0 or
light signals sent from them can reach O. The second class, the interior of the cone,
consists of points with timelike separation from point 0 (Lir > 0). For every
of these points, either it can be reached by a signal (such as a moving body) sent
from 0 with a velocity smaller than the velocity of light, e, or such a signal sent
from them can reach O. The set of these points belonging to said classes which can
be reached by signals sent from 0 is called the future of the point o. The past of the
point 0 is formed by these points belonging to the above classes from which a signal
can be sent to the point o. The third class consists of points with spaeelike
separation from the point 0 (Lir < 0). For every point, which belongs to this class,
there is an inertial frame in which this point and the point 0 are simultaneous.
Events l2 located at the points of the third class, cannot stand in any causal
relationship to events located at 0, because there exist no faster-than-light signals,
and thus no causal signal can come from any of these points to the point 0 and no
causal signal can come from 0 to any of these points. Events located at spacetime
points, which belong to the past and to the future of the point 0 (and only these
events), can stand in causal relationships to events located at the point O.
Let us consider the clock at rest with respect to the reference frame
S' (LIr"' = 0), which moves with a velocity v relative to a reference frame S. Let us
calculate the interval LiTfirst in the frame S' and then in the frame S
It follows from the formula (6), that the interval LiT coincides (up to the
coefficient e) with the time measured by the clock moving together with the frame
S'. Therefore it is called the proper time. If we compare the formulas (6) and (7) we
obtain that
(8)
which means that the time Lit' measured by the clock moving relative to S is shorter
than the time Lit measured by clocks at rest relative to S. This phenomenon is known
as time dilation. 13
The history of any point particle forms the so-called worldline of this particle in
spacetime. The length of this line, measured by means of the interval Li ... is equal
to l4 :
t t
Je"; 1 - v 2 / e 2 dt' S Jedt' (10)
to to
This inequality explains the weIl-known paradox of twins; the proper time
of the twin that moves and is subjected to accelerations is shorter than the proper
time of the other twin.
The dynamic equation of motion in STR, the counterpart of Newton's second
law, has the form
(11)
(12)
where r ijk - the flat affine connection. To repeat, flatness of affine connection
means that there exist global coordinate systems in which r~k=O. Such coordinate
systems are just inertial frames. The above mentioned pseudo forces are included
in the second term on the right-hand side of the equation (12). Because there is no
possibility in STR to connect the inertial (or affine) structure with the global mass
distribution in the Universe, we must connect it with aspacetime itself and thus we
must interpret the above theory of motion as an absolute one.
As in the case of Galilean relativity principle, in order to formulate precisely the
special principle of relativity we must express it in the language of absolute objects
and their symmetry groups. In STR we have the following absolute objects: the
affine connection r ~k and the metric gij.15 Their symmetry group is of course
(POINC). The special principle of relativity teIls that the group (POINC) is the
symmetry group of STR (M, r ijk , gij ). These symmetries inform us of such
important properties of spacetime as its homogeneity (expressed by the invariance
of the absolute objects of STR under the spatiotemporal translations), the isotropy of
space (expressed by invariance of these objects under the spatial rotation), the
isotropy of spacetime (expressed by their invariance under the Lorentz group) and
the symmetry in respect of mirror image reflection. As it has been pointed out
MOTION, SPACE, TIME 13
(13)
14 JERZY GOLOSZ
(14)
(15)
Did Earman prove that there could be no relational relativistic theory of motion?
The reply to this question depends on what we mean by a relativistic theory and its
spacetime. If by relativistic spacetime Earman means spacetime of GTR, his claim
that relativistic spacetime is irreconciliable with a relation al conception of motion
would only tell that Einstein's theory of relativity cannot be interpreted relationally
and it would be a counterpart of the classical assertion that the Newtonian theory
cannot be interpreted relationally. But if by a relativistic spacetime Earman means
the spacetime of whichever theory that would apply to the whole range of possible
velocities then he made a petitio principi error, because he did not prove that the
spacetime of every such a theory has to have the structure of spacetime of GTR.
It seems moreover, that such a proof cannot be established because we cannot
foresee what properties all potential theories of motion can have, just as we could
16 JERZY GOt.OSZ
not foresee the properties of the theory of relativity on the basis of Newtonian
theory. Earman may be right when he says that every adequate theory of motion has
to be absolute, but he did not prove that.
SUB Spacetime points are individuals and spacetime is a set of such points. 24
lassurne here the point of view of scientific realism which holds that one is
committed to believe in the existence of those entities that are ineliminably referred to
in one's best scientific theories.
The absolute-relational ontological controversy with respect to the status of
spacetime is not a dichotomy. There are two alternative kinds of antisubstantival
conceptions: relationism and property view. According to property view, points of
spacetime should be interpreted as properties of spatiotemporal positions of objects.
Relationism assumes that all spatiotemporal predications are relational (i.e. they refer to
the respective at least dyadic relations among bodies or events). The property view
takes something from either of the other positions: it agrees with relationism in denying
spacetime substantivalism and it agrees with substantivalism in recognizing irreducible
monadic properties of spatiotemporal position of objects.
Traditionally, the following relation between (REL) and (SUB) is assumed:
This view is based on the presupposition, that if motion is absolute, it must take
place with respect to a substantival space. If we assume both (16) and the
absoluteness of motion, we can conclude by modus ponens that spacetime is a
substance (SUB). Because spacetime substantivalism is denied both by relationism
and the property view, we could settle the ontological controversy in favour of
substantivalism.
This inference is criticized by Sklar (1976, pp. 229-232). His idea is to deny that
motion is relational (-REL), but at the same time to reject spacetime substantivalism
(-SUB). This position can be accepted, according to Sklar, if one admits that
acceleration is absolute, but at the same time it must be treated as a primitive,
monadic property of particles. Usually it is assumed that acceleration of a body is an
acceleration relative to something, for example, to other bodies, to fixed stars or to
inertial reference frames. Sklar proposes to treat the expression 'A is absolutely
accelerated' as a complete assertion, analogous to 'A is red'. He does not justify this
MOTION, SPACE, TIME 17
proposal, simply admitting that he can offer no explanation why some, systems
suffer no inertial forces whereas others do.
Sklar's idea was appraised in different ways. For example, Hoefer and Ray
(1992, pp. 575, 579) treat it as speculative and Teller (1991, p. 370) criticizes it as
ad hoc. On the other hand it is accepted by Friedman and, under some conditions, by
Earman. Unlike Sklar, Friedman (1983, pp. 232-236) attributes the primitive
property of absolute acceleration not to bodies, but to their trajectories. However, as
pointed out by Earman (I 989b, pp.l63-166), Friedman has not provided a
constructive alternative for the existing, absolute theories of motion, an alternative
which would be something more than instrumental exploitation of the existing
theories.
A more sophisticated interpretation of Sklar's idea is proposed by Earman
(l989b, pp. 126-128, 154, 214). Earman is of opinion that Sklar's idea is only
a 'very clever conjuring trick' (l989b, p. 214, n. 10). On the other hand, he hopes
that it can be developed so as to become acceptable. Specifically he proposes a so
called representational ploy,25 consisting in the assumption that physical reality
is basically relational and thus it should be described by relational physical theories.
Tbe substantival pictures of reality, as provided by extant theories, are only
representations of reality but the representation relation is one-many, so that many
(uncountably many) substantival pictures correspond to the same relational reality.
Tbe well-known Leibnizian argument against substantivalism can be interpreted in
the spirit of this reasoning. According to that argument, one and the same physical
system can be described by the substantivalist in different ways, as he introduces
a fictitious spacetime and then 'repositions' and 'reorients' this system in many
ways in spacetime.
Regarding the problem of motion, Earman's representational ploy consists
in treating absolute acceleration, which appears in Galilean spacetime,
as a representation of Sklar's primitive absolute acceleration. Tbe relationist cannot
treat the Newtonian or Einsteinian theory of motion as anything but a convenient
fiction, because they introduce affine (or inertial) structure of spacetime and thus are
substantival. Accordingly, he should try to use Sklar's notion of primitive absolute
acceleration in order to formulate a theory of motion which should be able to explain
and predict trajectories of particles. According to Earman (1989b, p. 128), such
a theory should include principles of motion which would be analogous to absolute
(Newtonian or Einsteinian) laws of motion, and these analogues must be close
enough, so that one could see that a given model of this new theory is represented by
each member of adefinite class of equivalent absolute models. At the same time, it
cannot introduce a vocabulary which fosters substantivalism. Earman is of opinion
that if such a theory is created, the existing absolute theories of motion could be
treated instrumentally and they would not imply that spacetime is a substance.
After submitting the above project of the representational ploy, Earman
announced (l989b, p.128) a partial realization of this project in the last part of his
work. Unfortunately, it is hard to find even a partial realization of his project in the
last chapters of his book. We find there only a representational ploy applied to GTR,
but it seems to be unsuccessful,26 and the problem of finding a theory of motion with
Sklar's absolute primitive acceleration is not discussed there at all.
18 JERZY Gowsz
Is not, however, the very possibility of existence of a theory, which would realize
Earman's representational ploy regarding the problem of motion, sufficient to
revoke (16)? The question is pointless, because it can be proved that such a theory
cannot exist. To be accepted in physics, a theory of motion has to enable one to
quantitatively describe the phenomenon of motion. For example, it has to enable
one to attribute to particles definite positions and velocities, which are essential in
calculating energy and momentum. In such a theory acceleration should be
expressed by definite number, just as it is in Newtonian theory, where acceleration is
calculated with respect to the class of inertial frames. We cannot be satisfied with
a purely qualitative theory, stating the existence of an absolute acceleration.
In a theory, in which acceleration would be treated as a primitive monadic property
of particles, it could not be attributed a numerical value because there would be
nothing with respect to which this value can be calculated. Thus Sklar' s proposal is
nothing but 'clever conjuring trick'. It is worth mentioning that Sklar's idea is
inconsistent with a fundamental property of motion, called by physicist relativity of
motion, according to which motion should always be related to a reference frame.
Thus the validity of (16) and absoluteness of motion should be treated
as an important argument in favor of substantivalism. However, this does not leave
a relationist in a hopeless position, because it has not been proved so far that
a relational theory of motion cannot be created. The relationist still can look for
a theory of motion consistent with (REL) or for a more general, non-substantival
physical theory which would imply relational theory of motion in agreement with
the equivalent to (16) formula:
The obj.ection raised lately against substantivalism, referring to the so-called hole
argument, 7 claims that substantivalism cannot be reconciled with determinism.
This objection, if valid, would reveal that theories, which are commonly believed to
be deterrninistic, like Newtonian mechanics or theory of relativity, are
indeterministic. However, as I show in another paper, the hole argument cannot be
employed with respect to the version of substantivalism, where metrical andlor
absolute (invariant under symmetry transformations) properties of spacetime points
are assumed to be their essential properties. If one takes into account restrictions
which are imposed by this assumption of essentiality on the active interpretation of
general covariance, the hole argument turns out to be invalid. 28
5. CONCLUSIONS
In this paper I tried to show what we can find out about spacetime by analyzing the
phenomenon of motion. Problems I was especially interested in were spacetime
symmetries, spacetime structures and the ontological status of spacetime. These
problems were discussed on the basis of the classical theories of motion contained in
Newtonian physics, special and general theory of relativity. To sum up, spacetime
symmetries of the above mentioned theories are (respectively) Galilean
transformations, Poincare transformations and the most general group of all
MOTION, SPACE, TIME 19
Jerzy Golosz
Institute oJ Philosophy,
Jagiellonian University
Grodzka 52,31-044 Cracow. Poland
E-mail: zbgolosz@kinga.cyJ-kr.edu.pl
NOTES
I would like to thank Prof. Helena Eilstein for inspiration and helpful comments.
1 SubstantivaIism. relationism and property view will be defined more precisely nmher.
, Though Leibniz was an ontological relationist he seems. however paradoxically. to accept that there
is absolute motion. Indeed. he wrote in his fifth letter to Clarke: 'However, I grant there is a difference
between an absolute true motion of a body and a mere relative change of its situation with respect to
another body. For when the immediate cause of the change is in the body, that body is truly in motion.
and then the situation of other bodies with respect to it will be changed consequently, though the cause
of that change be not in them.' (Loemker 1969, p. 706).
3 A body is in free motion when no external influences act upon it. See, for example, Kopczynski,
Trautman 1992, p. 26.
4 E. g. Reichenbach (1957) believes, that Mach's explanation of Newton's bucket experiment is equally
plausible as the Newtonian explanation.
5 See e.g. Friedman 1973, Kopczynski, Trautman 1992, Earman 1989b, Heller 1993.
6 A diffeomorphism 'I' is a bijection such that 'I' and '1'.1 are continuously differentiable (any number
of times). 'I'*Ai denotes the dragging aIong induced by the mapping 'I'of the geometric object Ai. For more
details see, for example, Friedman 1973, 1983.
7 The conservation law which corresponds to the symmetry in respect of mirror image reflection is the
parity conservation law in the quantum mechanics. This law has no equivaIent in the c1assical physics.
The parity is not conserved in weak interactions. See, for example. Crawford et al. 1957.
8 I will discuss in short two papers of Barbour (1974) and Barbour. Bertotti (1977), that are good
examples of the method used by the authors. This method was later developed in other papers.
9 When the symmetries become narrower. the list of invariants increases.
10 I follow here Earman (l989b). Barbour and Bertotti (1977) make use of different terminology:
the Machian symmetry mappings are called by them Leibniz group.
11 The Poincare transformations group is a composition of the narrower - and more known - Lorentz
transformations group with three-dimensional rotations. spatiotemporal translations and reflections.
The Lorentz transformations have the form I' = (I - vx!Cy.Jl-v'/c'. x' = (x - VI) I.JI-v'/c'.
y'=y. z'=z
12 By events I understand proper events, i.e. happenings. The term 'event' is used by physicist ambiguously,
to denote both proper events, i.e. happenings, and event locations (spacetime points).
20 JERZY GoWSZ
13 Tbe second well-known effect appearing in the STR is the so-called Lorentz-Fitzgemld length
contmction. Let us consider the rod of length 10 =<1x' in the frame X'Y'Z', in which it is at !'eSt. It follows
from the second formula of Lorentz transformation (ftn. 11) that <1x' = (<1x - vLlt) I.JI-V 2 /C 2 •
Tbe length I of the rod in the frame XJIZ is defined by <1x. assuming that LIt = O.
Hence I = lo.Jl- v 2 / c 2 •
14 See, for example. Kopczynski, Tmutman 1992, pp. 60-61.
15 Tbe metric and the affine connection are not independent; the former determines the latter. See, for
example, Kopczyitski, Tmutman 1992, p. 115, Friedman 1983, pp. 178, 355.
16 See Einstein 1949.
17 Oe Sitter found his solution for field equations with the cosmological constant: R'j - (1/2) g'j R = (81lG
/c' ) T,j - A gij, where A is the cosmological constant, which describes the hypothetical forces presumed
to act between astronomical objects. Einstein introduced the cosmological constant to his equations in
1917 in order to save the static model ofthe Universe.
18 See, forexample, Einstein 1949,lnfeld, Plebaitski 1960, Kopczyitski, Trautman 1992. Wald 1984.
I' See, for example. Infeld, Plebaitski 1960, Chaps. 2 and 3.
20 See, for example, Friedman 1983, Chaps 2 and 5.
21 X. Y €I M are causally connectible just in case there is a smooth causaI curve (i.e. a smooth curve with
causal tangent field gi! ~ ~ ~O) joining x and y. See, for example. Malament 1985, p. 617, Hawking.
Ellis 1973, p. 103, Heller 1991.
22 Tbe metrics gab and g'a b are conformaIly equivalent just in case there exists a smooth mapping t/1: M
~R such that g'ab= tfJ2 gab.
23 Earman 1989b, p. 102. See also 1989b p. 108, 1989a p. 85. It is worth noticing that Malament restriets
his considemtions to GTR.
24 In my (1999) I discuss the problem how to distinguish individuals from properties. See also
Augustynek 1994. If we assume that parts of spacetime are not points, but some extended objects. we have
to replace spacetime points in the above consideration by these objects.
25 See Earman 1989b, pp. 120. 127-128, 170-171.
26 Tbe Earman's representational ploy for GTR is based on Geroch's (1972) idea to coostruct a new version
of GTR without spacetime points - aversion expressed in the language of Einstein's algebras. See my
1999,2000.
27 See Earman and Norton 1987, Earman 1989b.
28 See my 2000.
REFERENCES
Augustynek, Z. (1994). Z ontologii czasoprzestrzeni. Filozojia nauki 6, 5-13.
Barbour, J. B. (1974). Relative - Distance Machian Theories. Nature 249,328-329, erratum Nature 250,
606.
Barbour, J. B., Bertotti, B (1977). Gravity and Inertia in a Machian Fmmework. Nuovo Cimento 38B,
1-27.
Berkeley, G. (1752). De Motu. In: D. M. Jesseph (ed. and trans.) De Motu and The Analyst. Dordrecht:
Kluwer Academic Publishers (1992).
Crawford, F. S., Cresti, M .• Good, M. L, Gottstein, K., Lyman. E. M., Solmitz. F. T., Stevenson, M. L,
Ticho. H. K. (1957). Oetection of Parity Nonconservation in A Decay. Physical Review 108, 1102-
1103.
Earman, J. (1989a). Remarks on Relational Tbeories of Motion. Canadian Journal of Philosophy 19, 83-
87.
Earman, J. (1989b). World &augh and Space-Time. Cambridge: MIT Press.
Earman, J., Norton, J. (1987). What Price Space-Time Substantivalism? Tbe Hole Story. British Journal
for the Philosophy of Science. 38. 515-525.
Einstein, A. (1949). Autobograjical Notes. In P. A. Schlipp (ed.), Albert Einstein: Philosopher
- Scientist. Evanstone IIIinois: North Western University Press.
Friedman, M. (1973). Relativity Principles, Absolute Objects, and Symmetry Group. In P. Suppes (ed.),
Space. Time, and Geometry. Dordrecht: D. Reidel.
Friedman, M. (1983). Foundation of Space-Time Theories. Princeton: Princeton University Press.
MOTION, SPACE, TIME 21
Galileo, G. (1632). Dialogue Concerning the ChiefWorld Systems - Ptolemaic & Copernican. (Trans.
by S. Drake), Berkeley: University of Califomia Press (1967).
Geroch, R. (1972). Einstein Aigebras. Communication in Mathematical Physics 26, 271-279.
Golosz, J. (1999). On Field's Argument for Substantivalism. International Studies in the Philosophy
ofScience 1,5-15.
Golosz, J. (2000). 0 tzw. argumencie dziury. FilozoJia Nauki 1, 35-72.
Hawking, S.W., Ellis. G.F.R. (1973). The Large Scale Structure of Space-Time. Cambridge: Cambridge
University Press.
HeUer, M. (1991). Osobliwy WszechSwiat. Warszawa: PWN.
HeUer, M. (1993). Fizyka ruchu i czasoprzestrzeni. Warszawa: PWN.
Hoefer. C., Ray, C. (1992). Review ofEarman (1989). British Journalfor the Philosophy ofScience 43,
573-580.
Horwich, P. (1978). On the existence ofTimes, Space, and Space-Times. Nous 12, 396-419.
Infeld, L., Plebanski, J. (1960). Motion and Relativity. Oxford - Warszawa: Pergamon Press - PWN.
Kopczynski. W., Trautman, A. (1992). Spacetime and Gravitation. Warszawa - Chichester: PWN
-John Wiley.
Loemker, L. E. (ed. and trans.) (1969). Leibniz, Philosophical Papers and Letters. Dordrecht: D. Reidel.
Mach. E. (1883). Die Mechanik in ihrer Entwiklung. 9 Auflage, Leipzig ,1993.
Malament, D. (1985) A Modest Remark about Reichenbach, Rotation and Relativity. Philosophy
of Science 52, 615-620.
Newton, I. (16681). De Gravitatione. In Hall, A. R., Hall M. B. (eds.). Unpublished Scientific Papers of
Isaac Newton. Cambridge: Cambridge University Press (1962).
Newton, I. (1729). Mathematical Principles of Natural Philosophy. (Trans. by A. Motte), Berkeley:
University of Califomia Press (1947).
Reichenbach. H. (1957). The Philosophy of Space and Time. New York: Dover.
SkIar, L. (1976). Space, Time and Spacetime. Berkeley: University of CaIifomia Press.
Teller, P. (1991). Substance. Relations and Arguments about the Nature of Space-Time.
The Philosophical Review, 3 Vol. C, 363-397.
Wald. R. M. (1984). General Relativity. Chicago: University ofChicago Press.
LESZEK M. SOKOLOWSKI
Abstract. Search for a unification of Einstein's general relativity and quantum theory in quantum gravity
theory is the most important and most difficult research programme in fundamental physics since early
1980's. In this artic1e we present motivation for the unification and sorne of main difficulties which
prevent contemporary physicists from achieving the aim. We focus our attention on the conceptual
problems regarding the notions of time, space and spacetime in quantum gravity. After explaining of why
gravity should be quantized (section 1) we present the modem notions of time and spacetime according to
general relativity (in section 2) and then compare them in section 3 with the concepts of time and space
as they are necessary in quantum mechanies in order for the probabilistic interpretation of the theory
to work. Since the concepts in both the theories are different and reconciling them in quantum gravity
is extremely hard, we present in section 4 attempts to achieve a less ambitious ailn. a fully quantum
theory of matter in a c1assical gravitational field. The current idea that the conceptual problems
of foundations of quantum theory may be solved with the help of quantum theory of the Uni verse
is discussed in the last section.
Keywords: relativity theory - quantum theory - time - spacetime - quantum gravity - quantum
cosmology.
23
H. Eilstein (ed.), A Collection 0/ Polish Works on Philosophical Problems
o/Time and Spacetime, 23-46.
© 2002 Kluwer Academic Publishers.
24 LESZEK M. SOKOWWSKI
Physicists seek for a theory of quantum gravity because they deeply believe in unity
of the Nature and the whole history of physics supports that belief. Gravity
is a universal property of matter since all forms of matter (all particles and fields)
interact gravitationally - in practice a material object may be defined as one that
is both affected by gravitational forces of other objects and exerts such forces upon
all of them, while the three other fundamental interactions of matter, Le.
electromagnetic, weak and strong ones, are only exerted by specific species
of elementary particles. There is only one more feature of matter that is universal:
an matter obeys quantum-mechanical laws. Empirically it is beyond any doubt that
all microscopic systems have quantum nature. Whether or not macroscopic bodies
obey quantum laws, i.e. whether there is a fundamentally classic (non quantum-
mechanical) world, is a subtle issue and the debate on it has not been completed.
As 1 will show below, there are strong arguments in favor of the universal validity
of quantum laws for all physical objects, even those which are traditionally regarded
as classical, and this standpoint is now shared by most physicists. The conviction
about the physical unity of the Nature makes it inconceivable to assume that gravity
might be an exception, that the gravitation al interaction (being a fundamental one)
might be devoid of quantum structure and be the only strictly classical entity
in the world. Of course, if gravity is not a fundamental (that is, elementary) force
and is composed of other forces, then it is not so evident that it should be quantized.
For instance one should not quantize the van der Waals forces acting between
molecules. Nobody, however, has been successful in decomposing gravity into other
known or unknown forces. Furthermore, even if gravity is a composite entity
described by Einstein's General Relativity, there are convincing arguments showing
that the picture of quantum matter generating a smooth classical gravitational field
may be inconsistent. It seems thus unavoidable to quantize gravity, i.e. to replace the
physical picture of the gravitational field as a smooth entity filling the space by that
of a swarm of discrete quanta of the field, named gravitons. This picture arises from
a strict analogy between gravity and electromagnetism: by quantizing the classical
electromagnetic field one assigns elementary particles (being quantum excitations
of the field) to it - photons; one expects that a similar effect occurs in the case
of gravity. Actually this analogy may be, as we shall see, misleading or even false,
but for the moment one may use it to visualize what is meant by quantum
gravitational field.
It is not surprising that quantum nature of gravity, if real, remains concealed
in all known laboratory and astronomical effects. In all these effects gravitational
forces are weak, we have not observed yet any process where gravity is strong.
It is likely that even in the regime of strong gravity it will not be easy to discern
its quantum structure. By analogy: it is rather hard to show that an intense bearn
of light actually consists of photons and it took several years in the beginning of 20th
century to prove it. Furthermore, physicists are not quite sure of how to recognise
quantum effects in a strong gravitational field (besides showing that the field
is actually discontinuous and is made of gravitons - and this picture, as mentioned
above, may be false).
It is currently believed that the domain of quantum gravity is determined by
the Planck scale, Le. quantum effects should become conspicuous for gravitational
THE PROBLEM OF TIME IN QUANTUM GRAVITY 25
processes occurring in the scale of Planck units. To describe quantum gravity one
needs to use three fundamental constants ofphysics: Newton's gravitational constant
G indicating that the interaction is gravitational, light velocity c since the interaction
is relativistic and Planck constant h responsible for its quantum nature. Out of these
3 constants one can define quantities bearing dimension of any physical quantity
by taking appropriate powers of the constants. These quantities serve as the natural
system of units for fundamental physics. The traditional units of length, time and
mass, namely the meter, the second and the gram, were introduced on purely
historical grounds and have no motivation in fundamental physics. The Planck units
made of h, c and G, when applied to ordinary atomic and nuclear physics, are either
very small or extremely huge, indicating that they determine the natural scale only
for quantum effects including gravity. Gravity, however, is negligibly weak
in phenomena studied by ordinary quantum physics. Consider a collision of two
electrons, each carrying relativistic energy of order of Planck energy E p = (hc5/GyJl2
= 101 9 GeV; for elementary particles it is a huge energy as it equals to the kinetic
energy of an airliner of mass 60 tons flying at velocity 900 km/h. For quantum
gravity effects, however, it is insufficient to have Planck energy in a collision:
disastrous collisions of airliners have occurred but no quantum effects ever
appeared. To this aim the colliding particles must ~proach each other at a very
small distance of order of Planck length L p =(hG/c3yJ = ur33 cm, that is 1020 times
smaller than the proton diameter. This means that only point particles (like electrons
or photons) are relevant, all extended objects like protons or whole atoms will not
do. One may therefore expect that quantum gravity will dominate over classical
(i.e. non-quantum) gravity for electrons with energy of order E p colliding 'head-on'
(at the moment of closest approach their distance should not much exceed L p).
This clearly means that quantum gravity is very exotic for the known physics.
In fact, using modern technology, to accelerate an electron to an energy comparable
with E p would need an accelerator of the size of the Milky Way! And there are no
such energetic particles arriving to the Earth in the cosmic radiation. At present
we can imagine only one place in the Uni verse where quantum gravity may actually
be essential: in a vicinity of aspacetime curvature singularity, i.e. in the very early
Uni verse just after the Big Bang and deep inside black holes, far below their event
horizons. These regions are inaccessible for observations.
Since there is no empirical evidence at all for quantum gravity, its theory must be
constructed on purely conceptual, theoretical grounds and internal consistency and
mathematical elegance are the only guides for the theoreticians. Although
the quantum structure of matter is regarded as the underlying one, physicists are
unable to construct a quantum theory of some processes ab initio. A classical theory
of the processes is essential at the outset. A procedure termed 'quantization'
is applied to that classical theory which transforms it into aquanturn theory.
Ordinary (Le. non-relativistic) quantum mechanies arises by quantizing classical
mechanics and quantum electrodynamics arises by quantizing classical Maxwell
theory of electromagnetism. If the physical object is a field rather than a mechanical
system, there exists a generic scheme of quantizing a classical field theory, named
quantum field theory, which correctly works for all known fields in the Nature
- with the exception of gravity. In most approaches to quantum gravity Einsteinian
26 LESZEK M. SOKOLOWSKI
General Relativity (GR) is used as a starting point. In the 1990's it became very
popular to replace GR by string theory; according to it the elementary objects are not
point particles but very tiny loops (closed strings). Vibrations (or excitations) of a
string appear as fundamental interactions and one of the vibrations represents
gravity (described by Einstein field equations of GR). In this way gravity gets united
with other interactions. At the end of 20th century most researchers were attempting
to quantize gravity via quantizing string theory. However string theory has many
hard problems and relativists criticise some of its assumptions. Relativists believe
that GR is a more appropriate theory to be quantized.
By definition, quantum gravity should be a consistent union of GR (or possibly
another classical theory of gravity) and quantum field theory (the latter is meant as
a procedure of quantizing a given classical field theory). These two theories are the
cornerstones of all physics and finding out their harmonious union would be highly
desirable. However they are very distinct. Quantum field theory (QFT) describes
elementary particles in their interactions (besides gravity) as material objects
existing in a given fixed spacetime - spacetime of Special Relativity (SR).
This spacetime is mathematically described by Minkowski space and I will identify
the two notions. Minkowski space acts as a rigid stage upon which all physics
is performed (physicists say that it is a 'fixed background' for any physical process)
and it remains insensitive to any process occurring in it. This is a very good
approximation to the genuine spacetime of the world around uso According to GR,
however, in the presence of gravity the stage is no more fixed and rigid. It should be
identified with the gravitational field itself and becomes a variable entity govemed
by dynamical laws, analogously to its matter content. Physical processes include
both matter and spacetime, since matter interacts gravitationally and gravitational
field is spacetime. The influence is mutual: matter affects spacetime and spacetime
affects matter. In GR there is no fixed background with respect to which one
describes dynamical processes; everything is dynamical and cannot be prescribed
from the outset. This picture of the stage actively reacting to what is occurring on
it is appropriate in the realm of strong gravitational fields; otherwise the
approximation of SR is entirely sufficient. One then sees that QFT and GR deal with
different things. What is more important, QFT is based on SR. As I will discuss in
more detail later, Minkowski space is essential for QFT. In particular, quantum
theory is based on the concept of time taken from SR. In some of its most essential
features it coincides with the Newtonian concept of time. This means that QFT and
GR in their standard formulations are incompatible. They become compatible and fit
together only in the limit of vanishing gravitation al interactions where GR becomes
trivial and is reduced to SR. The fundamental problem of quantum gravity may be
then stated as folIows: any theory oJ quantum gravity should describe extremely
strong gravitational fields as quantum effects while quantum theory is based on
concepts oJ time and space which are valid only Jor negligibly weak gravity.
From this incompatibility of the theories one infers that a theory of quantum gravity
(being a consistent unification of classical gravity and quantum theory) must be
radically noveI. The question arises which characteristic features of each of these
theories must vanish at the unification and what new concepts and categories will
TuE PROBLEM OF TIME IN QUANTUM GRA VITY 27
models and each model is a particular solution to Einstein field equations. The real
spacetime around us (or the spacetime of the entire Universe) should be described
by one of these models. Various models may substantially differ (the differences
may be greater than those between various electromagnetic fields, e.g. between
electromagnetic waves and the static Coulomb field), but these differences are
irrelevant for our considerations and we shall focus our attention on the common
features of all solutions to the gravitational field equations. These common features
form what hereafter will be referred to as 'spacetime model' or simply 'spacetime'
inGR.
The mathematical model of physical spacetime in GR arises from successively
overlapping four mathematical structures. The underlying structure or the basic
object is the spacetime in the sense of set theory, Le. a set of elementary events
named points. An elementary event is a primary, nondefinable concept. Intuitively
it is a phenomenon occurring in a moment at a point of space, such as a coJlision
of two elementary particles. It is usually written in introductory textbooks on
relativity theory that regarding an elementary event, one is not interested in what has
happened (the nature of the phenomenon) but only in 'when and where' it happened.
This statement is not quite precise, since the question 'when and where' already
presumes that the event has temporal and spatial coordinates while these cannot be
defined for an arbitrary set of points. The term 'event' does not imply that
something physical actually happens at a given spatiotemporal location - the
concept equally weil applies to these points in spacetime, where there is no matter
and nothing happens. The set of events is uncountable. The second structure
imposed on the set of events is that of a topological space: within the spacetime one
defines open and closed sets, the concepts of proximity, continuous maps and
topological maps (homeomorphisms). The concept of dimensionality of spacetime
does not appear yet on this level. It will arise after imposing the third structure
- that of a differential manifold. The latter means that locally each piece
of spacetime looks like a piece of an Euclidean space of some dimension.
The dimension of spacetime equals that of the corresponding Euclidean space.
A good two-dimensional model of a differential manifold is provided by any surface
(without self-intersections) in 3-dimensional Euclidean space such as plane, sphere,
cylinder, torus, ellipsoid, hyperboloid etc.
On this level one can introduce curvilinear coordinate systems in separated
pieces of the spacetime. These are generated by Cartesian (rectangular) coordinates
in the corresponding pieces of the Euclidean space. From a variety of experiments
one uniquely establishes that spacetime has four dimensions. This means that each
point (event) is identified by a quadrupie of numbers (coordinates) which
continuously vary between nearby points. However one cannot yet say that one
coordinate is temporal and the other three are spatial ones. The structure
of a differential manifold alJows one to define vectors on spacetime and differentiate
functions on it, but still there is no concept of the distance between nearby points.
These deficiences are removed by imposing the last, fourth structure - that
of (pseudo) Riemannian space, which is currently named Lorentz space.
This structure consists of the so--calJed fundamental metrlc tensor, which defines
a distance between nearby points. This Lorentz distance differs from the well-
THE PROBLEM OF TIME IN QUANTUM GRA VITY 29
known distance in any Euclidean space since it does not obey the triangle inequality
(any two sides of a triangle are together longer than the third side). Locally
geometry (and physics) in spacetime must be close to that in SR and in the latter the
Minkowski space distance is different from the Euclidean one. Now in general
spacetime is curved. The relationship of SR to GR is precisely that of a plane to an
arbitrarily curved (two--dimensional) surface to which the plane is tangent in a given
point. The metric tensor ('metric' for short) fully determines the geometry
of spacetime. I stress once again that while SR is a physically interpreted geometry
of Minkowski space, GR is a physically interpreted geometry of any Lorentz space.
From a mathematical viewpoint spacetime is a pair {differential manifold, metric}
where the manifold is made up of the three lower structures. According to GR the
differential manifold component of spacetime is fixed once for all (actually
this is not exactly so, but for the scope of this article we can neglect the subtleties)
and it is an 'absolute element' of the theory. Yet the metric is a physical
object - a dynamical quantity subject to some equations of motion (Einstein's field
equations). According to Einstein the metric and the spacetime curvature
(determined by the metric) are identified with the gravitational field which in this
way becomes geometrized. Metric (gravity) is the dynamical element of the theory.
Thus spacetime in GR consists of both absolute and dynarnical elements, while
in SR Minkowski space is entirely an absolute element of the theory. This difference
has profound consequences for quantum gravity.
One short remark. The hybrid nature of spacetime in GR is the source of the
fundamental problem in quantum gravity: what should actually be quantized? Most
researchers follow the division into absolute and dynamical parts of GR and
presume that one should quantize the metric field while keeping the other three
structures, Le. the differential manifold, fixed. In this way one is dealing with a field
theory on a differential manifold (without metric), while all methods of QFT have
been developed for a field theory on Minkowski space (which is flat Lorentz space);
it is unclear how to quantize in this case. Furthermore, there are conceptual
arguments showing that in quantum theory the division is different and the
differential manifold structure and the topological structure, and even the point set
structure also may be influenced by quantum effects (quantum fluctuations), and
thereby they are dynamical quantities. Nobody, however, has a convincing idea of
how to quantize these quantities and what is really fixed in quantum gravity.
Let us return to the classical spacetime of GR. Due to the Lorentzian nature of its
metric, all curves through a given point can be divided, in the vicinity of that point,
in the same way as in Minkowski space of SR, into 3 classes, according to whether
the square of their length (between any two points on a given curve) is either
positive or negative, or equals zero. In this respect the difference between a curved
and the flat spacetime lies only in the precise form of the formula for the square of
length (which is determined by the metric). Curves of zero length are called 'null
curves' and those for which the square of length is negative are termed 'spacelike
curves'. In Minkowski space all null curves emanating from one point form
a geometric figure which looks like a cone in Euclidean space and is therefore
named 'light cone' or 'null cone'. The same name is applied to the figure formed by
null curves emanating from one point in an arbitrary curved spacetime though it may
30 LESZEK M. SOKOLOWSKI
curves connecting two given points then in general their lengths will be different,
what physically means that the elocks travelling along these two worldlines will
measure different physical time intervals. This is quite obvious if one thinks in terms
of geometry: time is distance and distance is always measured along a chosen line.
In practice the distance between two cities on the Earth is often measured along
a highway connecting them, and this road is very seldom a straight line (more
precisely: a piece of the great cirele). And if the cities are connected by more than
one highway, one gets different distances between them.
If we were able to move with relativistic velocities (that is, ones elose to c)
or lived in a region of the Universe filled with strong and variable gravitation al
fields, the fact that each of us has its own time and that the age of a person depends
on his (or her) history (worldline) would be quite obvious. It is only due to the
weakness of Solar and cosmological gravity and slowness of our motions that we are
used erroneously to believing in a universal time, which equally flows for all of us
and for all physical objects around uso
The physical time has a direct operational meaning since it is measured by one
elock. The elock moves arbitrarily, suffers various accelerations, nevertheless by
definition it measures time intervals between points (events) Iying on its worldline.
The statement 'the elock measured 2 hours between the airliner took off from Paris
and landed in London' has a elear operational and geometrical (length of the elock's
worldline) meaning. Yet what is the meaning of the statement 'a laser light beam
was emitted in California and after a second it was reflected by a rnirror on the
Moon'? These two events cannot be connected by a timelike worldline since a dock
would then have to move together with the light beam and thus its worldline would
be null. The statement has no direct operational meaning. To determine the
corresponding time interval one should use two docks, one on the Earth and the
other on the Moon, and thus the difficult problem of their synchronisation arises.
The same holds for spacetime points which can be connected only by a (smooth)
spacelike curve. Stating that such two events are simultaneous is not operationally
grounded; to this end one must introduce some arbitrary definitions and procedures.
In fact, one can introduce the concept of the lapse of time for events wh ich are not
connected by a timelike curve, but only at the cost that this time is not physical - it
is deterrnined by differences in arbitrarily introduced time coordinates and I will
return to this problem below.
In Eudidean geometry as weil as in a curved spacetime one can define
distinguished lines - those which among all curves connecting two given points are
either the shortest ones (for Euelidean space) or the longest (in aspacetime). They
are called geodesic lines (obviously in Euelidean space these are straight lines) and
usually while speaking about the distance between two distant (in geometry 'distant'
means that the coordinates of given points differ by finite values rather than by
differentials) points one thinks about the length of a geodesic line connecting them.
One rnight therefore attempt to define the physical time not for arbitrary timelike
curves but solely for timelike geodesic lines. Such restriction, however, would be in
conflict with the physicist's everyday practice: in most cases docks and other
measuring devices are subject to various external forces and do not travel along
geodesic worldlines.
32 LESZEK M. SOKOWWSKI
The interval of the physical time between two given points depends both on the
timelike line connecting them and geometry (curvature) of the spacetime. Physically
the CUTvature is identified with gravity or, more precisely, with the changes
of gravitational forces between nearby points. They are narned 'tidal forces, in
analogy with the well-known forces producing tides on the Earth. It is often claimed
that time is affected by gravity, which makes it flow inequably. This is incorrect,
since time (as weil as space) cannot be separated from gravity. They are not two
independently existing entities, one of which might affect the other. The confusion
arises from the fact that both space and time have a very simple form in Minkowski
space where gravity is absent. It is more correct to say that time is an aspect
of gravity.
Above I focused on the physical time which has both direct operational and
geometrical meaning. Besides this concept of time which is physically most
important, there is also another concept of time - that of a coordinate time,
commonly termed 'time coordinate'. This is one of the fOUT coordinates of points
in spacetime. In the geometrical model of spacetime used in GR (the model consists
of all common features of various solutions to Einstein gravitational equations)
coordinate systems are introduced as folIows. The spacetime is divided into pieces
(as it was done while introducing the differential manifold structure) and for each
piece one separately chooses a coordinate system which can be interpreted as
coordinates in some physical reference frarne. Coordinate systems can be freely
changed in each piece independently of which coordinate systems are used in the
neighbouring pieces (there are some 'boundary conditions' which should be satisfied
by coordinate systems covering neighbouring pieces, but here we need not bother
about them); physically such changes of coordinate systems correspond to changes
of material reference frarnes extending over a given piece of spacetime. In general,
thus, a coordinate system is local, in the sense that except few cases (the most
important of them is the Cartesian coordinate system in SR) it covers only a piece
of spacetime though the piece may be quite large. In most cases physicists are
dealing with local problems and the question of whether our real spacetime can be
covered with aglobai coordinate system is irrelevant. Using the Lorentz metric of
spacetime the four coordinates split in 3 spatial ones and a temporal one. The time
coordinate also is narned 'time' by physicists, causing confusion and
misunderstanding among philosophers and laymen. Some physicists whose field of
interest is far from relativity also sometimes feel confused. The time coordinate has
a simple geometrical meaning: with the aid of it the given piece of spacetime
covered by the coordinate system is sliced into three-dimensional (hyper)surfaces
on which the time cordinate is constant. Each of these hypersurfaces is spacelike,
in the sense that every curve, which entirely lies on one of such hypersurfaces,
is a spacelike curve. In other terms these hypersurfaces are made (woven)
of spacelike lines; it is then plausible to regard them as physical 3-spaces. Since the
time coordinate is constant for all points of a spacelike hypersurface, one may
interpret such a surface as a set of simultaneous events. Choosing a coordinate
system in a region of spacetime one chooses a particular splitting of spacetime (in
that region) into time and space. Clearly one can split spacetime into space and time
in infinitely many different ways and none of them is physically (and geometrically)
THE PROBLEM OF TIME IN QUANTUM GRA VITY 33
existence of tachyons is not ruled out by Relativity Theory and whether they exist or
not is a matter of experiment or observation and not of theory (all known proofs that
they give rise to contradictions, such as the one given by Roger Penrose in his The
Emperor's New Mind are incomplete). Tachyons would possess bizarre properties.
If in one inertial frame a tachyon moves with some finite superluminal velocity, then
there is another frame in which it moves not only faster than light but also backward
in time, and a third frame where it travels at infinite velocity, and therefore
is present immediately at every point along its worldline. Tachyons' worldlines,
being spacelike lines, cannot in general be parametrized by a time variable.
For tachyons one would then have the effect opposite to that for ordinary matter
- 'universality of change of position': a tachyon can stand still in time but must
move in space. We assume that time flows because we see no effects caused by
tachyons.
Let us summarize this section: physical time is defined locally and is determined
separately for each timelike curve, it is the proper time for any object having this
curve as its worldline. This time is a geometrical entity being the length of
an interval of that curve. It may be measured by a clock associated with that object.
Physical space is a 3-dimensional set of points forming only spacelike curves.
This is also a local notion. Spacetime, at least locally, can be sliced into spaces
in infinitely many ways and none of them is physically or geometricaIly preferred.
These features are quite far from our intuition.
commuting ones can have at most a single common eigenvector. If a quantum object
is in astate which is an eigenvector of a set of commuting observables, the object
has definite values of a11 these quantities. Operationa11y this statement means that
these values of these observables will be found in appropriate measurements.
And since, as it was mentioned above, in general a quantum measurement destroys
a quantum state, to measure values of a11 commuting observables one needs to make
a number of simultaneous measurements. Once again, the simultaneity refers to the
preferred time. We conc1ude that the wave function depends on the preferred
physical time and 'feeIs' this time, while it is insensitive to any other concept of
time.
It is worth noticing that the state vector is aglobaI concept. Though at a given
instant of time one intuitively expects to find a given electron, say, one emitted in
a laboratory process, within the laboratory or in its c10se surrounding, the state
vector of that electron is a function extended (i.e. it is different from zero) over the
entire infinite physical space, and in order to establish the physical content of the
state vector one must know its value at every point of space. If e.g. a free electron
whose state was prepared on the Earth is such that it has adefinite value of
momentum (and energy), then its state vector is represented by a plane wave
spreading out over space up to the most distant galaxies despite the fact that our
intuition suggests that the wave function should be vanishingly sma11 far from the
Earth. And for the outcome of experiments made on this electron in a laboratory on
the Earth, the shape and amplitude of its state vector in the farthest regions of the
Uni verse (whether or not it is still a plane wave) is significant. In general, every
quantum system is spatia11y nonlocal since its wave function in principle is extended
over the entire space; accordingly, the state vector provides useful physical
information about the system provided there exists a global time which is
independent of the dynamical evolution of quantum systems and defines a unique
physical simultaneity of events. (This dependence of the concept of the wave
function on the assumption of the absolute character of simultaneity has already
been mentioned above.) The existence of this preferred time does not fit the nature
of time in GR.
The problem of time in quantum gravity is now c1ear. On the one hand this
theory, as a quantum theory, requires a universal physical time, conceived as an
external parameter measuring any evolution and independent of it; on the other hand
in the specific case of quantum gravitational field time, being a geometrie aspect of
gravity, should be conceived as a dynamical variable to be quantized. In particular,
if one takes GR as a c1assical theory of gravity to be quantized (and the string theory
which is currently regarded by some researchers as an alternative to GR is in some
aspects deficient), then its concept of time is useless for the needs of quantum
theory.
Let us consider the concept of relativistic causality. Ordinary matter (massive
and rnassless partic1es) cannot move faster than light. Only if a light signal or some
other, slower signal, may connect two spatially distant events, then in principle the
physical situation in the later event may be influenced or determined by that what
occurred in the earlier event. (We are now not interested whether or not a signal was
actua11y emitted and affected the later event, but only in the possibility of causing
THE PROBLEM OF TIME IN QUANTUM GRA VITY 37
the influence.) In ordinary QFr, which is based on SR, one therefore introduces the
principle of relativistic causality (or 'causality condition'); if spacetime points x
and y are joined by a spacelike vector (and thus there exists an inertial frame in
which both points are simultaneous), then the values of any physical field or any
other physical quantity (e.g. the wave function) at points x and y are independent of
each other - are 'causally disconnected'. Clearly the principle is weIl founded.
Yet in a curved spacetime it depends on its metric whether the vector joining two
fixed points is spacelike or timelike. And in quantum gravity the metric is
a dynamical quantum object experiencing quantum fluctuations. Hence the length of
the vector also fluctuates and the fluctuations may convert in a chaotic way
a timelike vector into a spacelike one and vice versa. The very concept of classifying
vectors (and curves) in timelike, spacelike and null ones makes sense only for
a smooth metric field on the spacetime; in case of discontinuous chaotic fluctuations
of the metric one cannot say that a curve has adefinite length. Physicists say that the
quantum effects of gravity smear out the light cones (make them fuzzy) and one is
unable to establish with certainty if a given point lies inside or outside the cone of
another point. The concept of the distance between nearby points becomes
probabilistic and the same holds, in consequence, for the concept of the length of
any curve. The quantum nature of the metric assigns some probability to the
statement that the length of a given curve joining two fixed points has some definite
value. The length becomes a quantum observable. In general in quantum theory an
electron in a given state has no definite value of energy and its wave function
determines probabilities of measuring different values of energy - its state vector is
a superposition of eigenstates corresponding to these energy values. One imagines
that in quantum gravity in a similar way astate vector assigns probabilities for
a given curve between two points for various outcomes of the measurement of its
squared length. (I say 'imagine' because actually nobody is able to construct an
appropriate wave function for aspacetime conceived as a dynamical entity, though
quite recently there have appeared some promising results.) The outcome is that
at small distances (where the quantum effects should be most significant) the answer
to the question whether a given pair of points is causally connected, can be merely
probabilistic.
For the same reason there is only some probability for any value of the velocity
of any signal. Clearly such a 'probabilistic time' is unsuitable for parametrizing any
curve and thus an evolution of a physical object having this curve as its worldline
cannot be described in terms of this time. For the needs of quantum theory the
probabilistic time is completely useless.
I wish to emphasise on ce more that the whole conceptual and mathematical
structure of QFr is based on the assumption that all physical processes take place on
a fixed rigid stage - the flat spacetime (Minkowski space) which is insensitive to
what is occurring on it and thus is an absolute element of the theory. Time is an
essential element of this stage. Physicists are unable to construct a quantum theory
without the fixed stage and the external global time and in this sense quantum
mechanics is less perfect than GR. The beauty of Einstein's General Relativity,
which by many is regarded as the most perfect theory of physics, lies mainly in the
fact that it has so few absolute elements and in particular it has no fixed stage. GR is
38 LESZEK M. SOKOLOWSKI
curvature fluctuations become so violent that they are capable of tearing holes in
spacetime and changing its topology. An example of a change of topology is
provided by introducing a 'wormhole' on a plane. A wormhole is formed on aplane
by cutting two openings in it, stretching the cut edges into two tubes and then
joining them together (in mathematics the wormhole is often named 'a handle').
The topology of the plane with a wormhole is different from that of Euclidean plane.
By forming in the same way a wormhole on a sphere one converts the sphere into
a closed surface which has the topology of the torus. There are infinitely many
two--dimensional surfaces with different topologies. One can also construct an
analogous wormhole in aspacetime. Clearly the wealth of various topologies
for 4-dimensional spacetimes is much larger. Wheeler imagines that at the scale of
the atomic nucleus spacetime is quite smooth, at much smaller distances there
appears some roughness and at the Planck scale the curvature and topology of space
(he assurnes that spacetime can be somehow split into space and time) are
continually undergoing violent fluctuations. In this picture space is in astate of
perpetual turmoil; it is 'boiling', with wormholes and other more complicated
structures (all of order of Planck length) continually forming and disappearing.
Wheeler's picture is often called 'spacetime foam'. To explain it let us consider
a 2-dimensional case. Initially one has a plane. Quantum fluctuations form on it
a number of wormholes and surfaces of more complex topology. Further
spontaneous fluctuations appear not only on the 'free' parts of the plane, new objects
develop also on the already existing ones: wormholes generate wormholes and these
generate next wormholes etc. The initial plane quickly evolves into a complicated
structure and the process reminds the formation of a soap foam. (The object is
rapidly changing and at each instant its is similar to a sponge). This raises the
question concerning the true dimensionality of the space at the Planck scale. The
soap foam is actuallY an extremely complicated two-dimensional surface, but the
multitude of its topological connections gives it the appearance of a three-
dimensional object. It was therefore conjectured that actually space has only two
dimensions and this could be detected at the Planck scale while due to the
complexity of its topology space seems to have three dimensions (and to be smooth)
at larger distances.
This picture of space consisting of huge number of topologically and
geometrically complicated objects of the Planck size which rapidly arise, grow and
break or diminish and fade, so that small regions of space violently evolve (though
there are no long-term changes) is very impressive. Now, however, it is almost sure
that it is false. One objection is easy to present using a two-dimensional model of
space as a plane. A typical process for a wormhole would be to pinch off and leave
two 'dimpies' or 'pseudopods' on a plane. In general, if an elementary process takes
place, the inverse process should also be possible. Thus, two pseudopods should be
able to join and form a new wormhole. There are, however, some arguments to the
effect that while it is quite probable for a quantum wormhole to pinch off and
gradually disappear, the probability of the reverse process should be much smaller.
The wormholes would then be destroyed by quantum fluctuations rather than
formed; a similar fate would meet most of other quantum space structures. It is
extremely difficult to make a coherent model of topological transitions. Though the
40 LESZEK M. SOKOLOWSKI
practical importance, since it described rare and exotic effects. In the next decades
GR triumphantly entered in extragalactic astronomy and astrophysics. Black holes
and neutron stars have become ones of the most interesting and intensively
investigated objects in the Universe. In its study of the early Uni verse relativistic
cosmology has provided powerful means to test modem theories of elementary
particles. (It is sometimes expressed in the statement that the early Uni verse was the
most powerful accelerator which has ever existed). The success of the Standard
Model in particle physics (the unification of electromagnetic, weak and strong
interactions) meant the revival, on a new basis, of the old Einstein's idea of
unification of all fundamental forces of Nature and convinced most physicists that
gravity should also be unified in one theory with other interactions. In a famous
lecture at Cambridge in 1987 Steven Weinberg acknowledged GR as a foundation of
particle physics. Parallelly with the ascent of GR it has also been gradually
recognized by rnany scholars that the lasting troubles with understanding the
foundations of quantum theory are not of merely psychological nature but follow
from real conceptual problems. The criticism of the Copenhagen interpretation
raised by Einstein has been revived and became, after fifty years, again a matter of
a deep scientific debate. A number of interpretations of quantum mechanics,
different from the Copenhagen standard one, were developed and they are no more
regarded as insignificant curiosities. In 1970's Roger Penrose introduced a bold and
controversial conjecture that quantum mechanics is intirnately connected with
gravity and the reduction of the state vector actually is caused by gravitational
interactions.
To avoid any misunderstanding: quantum mechanics is the correct theory for
molecular, atomic and nuclear physics, confirmed in innumerable experiments and
its validity in these domains is beyond any doubt. There is no comeback to a 'good
classical' physics. All the criticism of quantum theory regards only its conceptual
foundations - its interpretation - and these issues cannot be solved by experiment.
The first objection is of heuristic nature and explains why is it possible at all to
criticise the Copenhagen interpretation without putting into doubts the entire
quantum mechanics. In all other physical theories besides quantum theory an
interpretation of a theory (indicating what the theory rnay and should actually
predict, what is measurable and which elements of the theory cannot be verified by
experiment) constitutes an inherent part of the theory. A change of interpretation
requires an appropriate change of the rest of the theory (both its physical content and
its mathematical forrnalism). In quantum mechanics, however, the Copenhagen
interpretation was imposed on the experimentally verifiable part of the theory and
the very fact that there are possible and do ex ist several different interpretations
which do not change the verifiable predictions of the theory, clearly shows that in
this case any interpretation is connected in a looser way to the dynamical part of the
theory.
Another objection indicates that the orthodox Copenhagen interpretation is fully
applicable to non-relativistic quantum mechanics of one or many (finite number)
particles while in the case of the most advanced part of quantum theory - quantum
field theory - it is applicable in some aspects only. As I mentioned above, the basic
concept for QFf is that of quantum vacuum state. QFf is probabilistic in the sense
THE PROBLEM OF TIME IN QUANTUM GRAVITY 43
problem is real and hard. Since quantum theory is nonseparable, any attempt to
establish boundaries separating a quantum system from the rest of the Universe is
merely an approximation which makes sense only in some situations. In reality there
are no many separate quantum systems - there exists only one quantum system
having no surrounding. It is the entire Uni verse. In reality quantum mechanics does
not apply to single electrons, atoms or molecules, as it was believed for more than
half a century, it applies to the entire Universe. Though it may sound astonishing, it
is an unavoidable consequence of the fact that the theory is applicable (in
approximation) to elementary particles. We arrive therefore at the conclusion that
the genuine core of quantum theory is quantum cosmology and that means that in an
adequate quantum description of Nature gravity must be taken into account. In 1990
M. Gell-Mann and J. Hartle wrote: 'Quantum mechanics is best and most
fundamentally understood in the framework of quantum cosmology'. And contrary
to a first impression the statement is quite plausible. In fact, all previous attempts to
give a fully satisfactory formulation of the conceptual foundations of quantum
theory based on our knowledge and understanding of quantum properties of
elementary particles and atoms have failed; there is no other way than to solve the
problem with the aid of quantum physics of the entire Uni verse.
General Relativity is the first theory capable of describing the entire Uni verse
(Newton's gravity theory is not) while all other theories deal with local phenomena.
In 1917 Einstein recognized that his theory includes some solutions to the
gravitational field equations representing the metric for the whole spacetime filled
with a continuous distribution of matter (the totality of stars and galaxies is
described as some kind of a homogeneous fluid). The set of cosmological solutions
to Einstein field equations forms the mathematical basis of relativistic cosmology.
These solutions determine the infinite set of such models of the Uni verse which are
consistent with GR. Of course, which cosmological solution actually describes the
real Universe should be deterrnined by observations. According to quantum theory
the Uni verse is the unique quantum system and quantum cosmology is a specific
part of quantum gravity theory in a way analogous to that relating classical
relativistic cosmology to GR. In quantum cosmology one assigns a wave function to
the whole Uni verse, and in order to recover all physical features of the system one
may formally apply to this state vector all the procedures applied in ordinary
quantum theory to the wave function of a rnicroscopic system like an atom.
However in the case of quantum cosmology the conceptual problems undermining
quantum gravity become terrifying. 1) Since the partition of the studied whole into
the quantum system and a classical observer (measuring instrument) disappears by
definition (everything that exists is included in the system under consideration), it is
unclear how to describe a measurement. 2) The probabilistic interpretation of the
wave function is doubtful, because when the object under study is the entire and
unique Uni verse, one is prevented from resorting to the concept of a statistical
ensemble. This is a significant novelty: up to now it has been always presumed in
physics, at least in principle, that the number of objects under study is unlimited.
In cosmology one has precisely one object to describe. In classical relativistic
cosmology the problem is resolved in the weIl known way: one has to fit to the
Uni verse one solution out of the infinite set of cosmological solutions to Einstein's
THE PROBLEM OF TIME IN QUANTUM GRA VITY 45
Leszek M. Sokolowski
Astronomical Observatory,
Jagiellonian University
Orla 171, 30-244 Cracow, Poland
E-mail: UFLSokol@TH.1F.UJ.edu.pl
MICHAL HELLER
Abstract. Basic ideas of noncommutative geometry are briefly presented. This mathematical theory,
being global from the very beginning, can be used to model physics in which local concepts, such
as those of time instant and space point, are meaningless. In spite of the lack of the standard time concept
a "noncommutative dynamics" can be defined. Noncommutative generalizations of causality, probability
and chance are discussed.
1. INTRODUCTION
47
H. Eilstein (ed.), A Collection of Polish Works on Philosophical Problems
ofTime and Spacetime, 47-56.
© 2002 Kluwer Academic Publishers.
48 MICHAL HELLER
3. NONLOCAL SPACES
Noncommutative geometry is indeed a powerful generalization of the standard
geometry. Many "pathological spaces" can be regarded as noncommutative spaces.
For example, those spaces which, from the topological point of view, are reduced to
a single point, turn out to be workable, noncommutative spaces when treated as pairs
(M, A) with A a suitable noncommutative algebra. This is due to a striking property
of noncommutative spaces - they are non-Iocal entities: the concepts of point and
its neighborhood are, in principle, meaningless in them.'
Functions on a manifold "feei" points. This fact is used in the definition of
function multiplication. Two functions,j and g, are multiplied by multiplying their
values at every point x, i.e., (f. g)(x) =ftx)·g(x). (Of course, this is the same as
g(x) . ftx) and, consequently, the multiplication of functions is commutative).
A given point x can be identified with the set of all these functions that vanish at x.
This set is called a maximal ideal of the given algebra C. In other words, points in
the manifold (M, C) can be identified with maximal ideals of the algebra C. In the
family of all algebras to have maximal ideals is an exception rather than a rule, lind
noncommutative algebras, in general, have no maximal ideals. This is why
noncommutative spaces are non-Iocal. They do not consist of points, and only
global concepts refer to them.
This is the radical change of perspective, which compels us to go outside the
range of validity of the usual set theory. For the expression "belonging to a set" to
have a meaning, there should be the possibility of identifying elements of the
considered collection by means of at most a denumerable family of properties.
In noncommutative geometry such a possibility does not exist, if we decide to use
only measurable maps between spaces (Connes 1994, p. 74); but to use
nonmeasurable maps would be as bad as going beyond the set theory.
4. TIMELESS DYNAMICS
To explore conceptual horizons that are open by noncommutative geometry let us
make the bold assumption that the fundamental level of physics is modeled by
a noncommutative geometry. Let us notice that in such a model there could be no
space and no time in their usual meaning since space consists of points and time
consists of instants. In fact, such a noncommutative model has been proposed
(Heller et al. 1997; Heller and Sasin 1999, 2000) but in the present study we shall
refer to it only for illustrative purposes. We treat the hypothesis that the fundamental
level is noncommutative in a purely heuristic way. It is supposed to help us
to realize the degree of generalization in passing from commutative geometry
to noncommutative geometry.
The fundamental level of physics "is situated" beyond the so-called Planck
threshold, which is characterized by the Planck length [PI = 1(J33 cm, and the Planck
time tpl = ]0-44 s. This threshold can be found "in two directions": firstly, if we go
backwards in time to the close vicinity of the Big Bang, when the "age of the
Universe" was of the order of 1044 s; secondly, if we go (now, i. e., at the present
cosmic epoch) deeper and deeper into the strata of the Uni verse until we reach
TIME AND PHYSICS 51
distances of the order of 10-33 cm. Since, however, on the strength of our hypothesis,
there is no space and time (in their usual meaning) beyond the Planck threshold,
both these directions (back in time and deeper in space) turn out to be the same
direction!
What could physics look like in the absence of space and time? When we think
about physics, we first of all think about dynamics, i. e., about physics of motion.
With no space and no time there can be no motion in the usual sense, but there can
be an authentie, albeit generalized, dynamics. To see this let us recall the following
facts.
In dynamical equations (for instance, in the Newtonian equations of motion)
time appears as a parameter whieh measures the change. However, instead of using
equations we can equivalently describe the dynamics with the help of vector fields.
In the usual setting such vector fields, called integral vector fie lds , consist of tangent
vectors to the trajectories of a given dynamical system. Let us also recall that
a vector is essentially a derivative of a function; for instance, the velocity vector is
the derivative of the distance function with respect to time. The concept of a vector
is a local concept, but the concept of a vector field has agiobai aspect that can be
generalized to the noncommutative setting.
Let (M, C) be a manifold. Then, formally speaking, a vector (tangent to M) is
a linear mapping
d: C-K:
satisfying the well-known Leibniz rule (whieh says how to differentiate the product
of two functions belonging to C). In noncommutative geometry, there is a non-Iocal
counterpart of the above concept that can be thought of as a generalization (and
a 'delocalization') of the vector field concept; it is called derivation of an algebra C,
and is defined to be a linear mapping
D: A-,lA
from a not necessarily commutative algebra A to itself, satisfying the Leibniz rule.
This concept can be used to define a generalized noncommutative dynamics. In this
way, the intuitive concept of motion, as a change of place in time, is replaced by an
abstract idea of mapping from an algebra A into itself which satisfies properties
analogous to those whieh are satisfied in the commutative case (linearity and the
Leibniz rule).
There are various noncommutative algebras, and by choosing the 'correct one'
we can more adequately model temporal properties of noncommutative dynamical
systems.
5. EMERGENCE OF TIME
The concept of state was used in physics for a long time. Although the state of
a given physical system can be represented as a point in aspace (calIed the phase
52 MICHAL HELLER
space), it has also global connotations. It is the whole physical system that can be in
this or in another state. In noncommutative geometry states appear naturally (as
functionals on the considered algebra, which are positive and with unit norm), and in
some cases 2 they can be regarded as substitutes of points.
In the rich family of noncommutative algebras especially important ones are the
so-called von Neumann algebras. Roughly speaking, a von Neumann algebra is
a CO-algebra3 together with a distinguished state. 4 We are not going to define
the CO-algebras5 ; for our purposes it is enough to state that a CO-algebra is an abstract
algebra having all essential properties of the algebra of observables of quantum
mechanics. All algebras considered in the present paper can be naturally made
CO-algebras. From our point of view, it is interesting that if A is a von Neumann
algebra, then the dynamies in the noncommutative space (M, A) can assume a rather
familiar form of a set of equations describing the evolution of the system with
respect to a certain parameter which imitates time. This is possible due to the elegant
Tomita-Takesaki theorem. Roughly speaking, this theorem states that if A is a von
Neumann algebra, then there is a farnily of mappings (automorphisms)
<lt: A-+A
forming a one-parameter group, called modular group, which depends on the state
on the algebra A.6 Let us comment on the possible physical meaning of this theorem.
The existence of the parameter t, indexing the family of mappings <lt: A -M, and the
fact that this farnily fonns a one-parameter group, allows one to use t as a parameter
describing a certain type of dynarnics, in the same sense as the usual time parameter
is used, for instance, in the Newtonian dynamies. There is, however, an important
difference: in the Tomita-Takesaki theorem the parameter t depends on the state on
the algebra A. Consequently, it can be interpreted as time, but a time dependent on
the state in which the considered system finds itselr.? If the considered physical
system goes over to another state, it also switches to another temporal regime.
However, it is interesting that the situation improves if we suitably restriet the set of
"adrnissible" automorphisms of A.
There is a subset U of the above algebra A satisfying the so-called unitarity
condition;8 the set U forms what is called the unitary group of the algebra A. With
the help of elements of this group one defines the equivalence relation, called inner
equivalence, in the set of all automorphisms <lt: A -+A. 9 The set of all equivalence
classes of this relation is called a group of outer automorphisms of A, and is denoted
by Out(A). In general the one-parameter group of the Tomita-Takesaki theorem does
not consist of inner equivalence automorphisms, but it can be constructed using only
elements of Out(A). In this case the one parameter group consists only of inner
equivalent mappings, and it can be demonstrated that it no longer depends on states
on the algebra A. In other words, if we suitably identify some mappings <lt: A-+A
(by using inner equivalence relation), the one-parameter groups 'improve' and we
obtain 'time parametrization' that does not depend on states. It is a remarkable
circumstance that this 'improved time' has been achieved with the help of unitary
properties of some elements of the algebra A. It could be related to the fact that the
TIME AND PHYSICS 53
ne ver move forward; on the other hand, viewing the world in the light of ever new
concepts could be regarded as the most conspicuous contribution of science to the
human culture. However, the truly fruitful conceptual evolution never is driven "by
itself'; it must be a strategic part of a struggle to solve a problem. In fact, there are no
isolated concepts and no isolated problems in science. Every change in the meaning
of a concept, every step forward in the process of a problem solving, results in shifts
of meanings of other concepts and opens new pathways leading toward new insights.
There are many mechanisms of this evolution. One of the most frequent and the most
fruitful of them is the strategy of generalizations.
In the foregoing sections we had an opportunity to consider the generalization
of several concepts in the process of transition from commutative geometry to
noncommutative geometry. In the present section I shall point out to some
consequences this process could have for philosophical thinking. I shall do this by
considering two examples, namely the generalizations of the concepts of causality and
of chance. There is no need to convince the reader that both these concepts play an
important role in shaping our views concerning many philosophically interesting
aspects of the world.
The problem of causality is notoriously difficult and has a long history
in philosophical debates. Here I shall restrict myself to some very simple remarks.
It seerns rather obvious that a cause and its effect must be two different events and that
a cause rnust precede its effects. As it is weIl known, there were attempts (Hume) to
reduce causality to the temporal order. According to these views b propter a can mean
nothing but b post a, and this relation holds in all similar cases. But in a
noncommutative world there are no separate events and, in general, there is no
temporal order. Does this mean that causality becomes an empty concept? It rather
must be suitably generalized. And it is the mathematical structure of noncornmutative
geometry that suggests how this should be done.
Spacetime of the standard theory of relativity carries what is called the causal
structure. It is defined by a Lorentz metric, which at each point of spacetime
determines the local light cone determining the pattern of propagation of causal
influences. When one changes to a noncommutative setting the causal structure
remains, but is deprived of all local aspects. 1O The pattern of causal interactions is no
longer defined by locallight cones but rather by some noncommutative counterparts of
vector fields (which are global entities), the same 'vector fields' that are responsible
for the dynamics of the system. The noncommutative generalization of causality
liberates this concept from its usual dependence on time and locality. It seems that the
essence of causality consists in a dynamical nexus rather than in the distinctness of
a cause and its effects and their temporal order.
We usually think about chance in terms of probabilities. If an event happens which
apriori had a small probability to happen, we are inclined to think that it either
happened by chance or was designed by an intelligent agent. Here we put aside the
problem of design or purpose and focus on chance. With the idea of chance we almost
automatically associate the feeling of expectation and uncertainty. What does remain
of this idea if we dissociate it from the notion of the flow of time and multiplicity?
As we have seen in the preceding section, the noncommutative counterpart of
probability is encoded in a suitable von Neumann algebra, and our intuitive ideas
TIME AND PHYSICS 55
Michal Heller
Faculty of Philosophy,
Pontifical Academy ofTheology
Franciszkanska 1, 31-004 Cracow, Poland
E-mail: mheller@wsd.tamow.pl
NOTES
I The phrase 'in principle' is important here. In some special cases a noncommutative space consists of
points, but in such cases points have an internal structure (see Masson 1996, p. 92). This never occurs
in commutative geometry. All noncommutative spaces considered in the present paper do not consist of
points.
2 If they are the so-called pure states.
J Read: C-star algebra.
4 This is not a definition. In the present essay we are satisfied with intuitive descriptions. The more
inquisitive reader should consult the specialliterature, for instance in Sunder (1987).
5 For the definition see, for instance, Sunder 1987, p. 14, or Madore 1999, p. 155.
6 For the precise formulation of the Tomita-Takesaki theorem see, for instance, the above quoted book
by Sunder (1987), chapter 2.
7 Dur interpretation of the Tomita-Takesaki theorem in terms of astate dependent time is based on the
following papers: Connes and Rovelli (1994) and Heller and Sasin (1998).
8 Which states that aa' =a'a =/, aE A.
9 For the definition consult works by Connes and Rovelli (1994) or Heller and Sasin (1998).
10 In Connes' approach to noncommutative geometry so far only Riemann metric has been defined, but in
other models also Lorentz metric can be made meaningful.
" This idea should be understood in the following way. There exist an infinite number of universes.
Every imaginable combination of physical constants, initial conditions, etc. is implemented in at least
one of them. We live in the highly ordered universe since only in such a universe Iife is possible.
56 MICHAL HELLER
REFERENCES
Connes, A. (1994). Noncommutative Geometry. New York-London: Academic Press.
Connes, A and Rovelli, C. (1994). Von Neumann Algebra Automorphisms and Time-Thermodynamics
Relation in GeneraIly Covariant Quantum Theories, Classical anti Quantum Gravity 11, 2899-2917.
Heller, M. and Sasin, W. (1999). Noncommutative Unification of General Relativity and Quantum
Mechanics,lntemational Journal olTheoretical Physics 38,1619-1622.
Heller, M. and Sasin, W. (1998). Emergence ofTime, Physics Letters A2SO, 48-54.
Heller, M. Sasin, W. and Lambert, D. (1997). Groupoid Approach to Noncommutative Quantization of
Gravity, Journal 01 Mathematical Physics 38, 5840-5855.
Heller, M., Sasin, W. and Odrzyg6ZdZ, Z. (2000). State Vector Reduction as a Shadow of
Noncommutative Dynamies, Journal 01 Mathematical Physics 41, 5168-5179.
Landi, G. (1997). An Introduction to Noncommutative Spaces and Their Geome/ries. Berlin-Heidelberg:
Springer.
Madore, J. (1999). An Introduction to Noncommutative DijJerentiat Geometry anti Its Physical
Applications. Cambridge: Cambridge University Press.
Manin, Y. I. (1991). Topics in Noncommutative Geometry. Princeton: Princeton University Press.
Masson, T. (1996). Geometrie non commutative et applications a la theorie des champs, Vienna, Preprint
of the Erwin Schrödinger International Institute for Mathematical Physics, no 296.
Smolin, L (1997). The üle ol/he Cosmos, New York-Oxford: Oxford University Press.
Sunder, V. S. (1987). An Invitation to von Neumann Aigebras. New York-Berlin: Springer.
JAN CZERNIAWSKI
The relativity theory and relativistic cosmology have served for a long time as the
main source of arguments against objectivity of time flow. Initially, the special
theory of relativity (STR) seemed more suitable for this purpose, as allegedly
inconsistent with the very notion of time flow. For time can flow only if in the
spacetime some relation of absolute simultaneity obtains. which corresponds to the
temporal becoming of events. But the Minkowskian spacetime does not contain such
a relation. The only relation of simultaneity defined in it in a natural way is relative.
i.e. dependent. in a sense. on the point of view'.
However. this argument would be effective only on condition that Minkowskian
spacetime represents the whole spacetime structure of the world. It is true that STR.
which is consistent with the results of all local experiments. ascribes to the physical
spacetime a symmetry. which rules out endowing it with any richer structure.
Nevertheless. there is no ground for assuming that physical measurements can grasp
all aspects of reality. Occam's razor would be effective in removing a non-physical
absolute simultaneity only if the postulate of its existence were not necessary for
explaining some aspect of our experience. No doubt temporality. not very
fortunately called theflow oftime. is such an aspect.
Thus. ascribing to the world aspacetime structure containing an absolute
simultaneity seems advisable. The physical spacetime would be a reduct2 of such
arieher structure. representing all the absolute spacetime relations that are
operationally definable. If the former were Minkowskian spacetime. the latter would
have to be at least as rich as the result of supplying it with absolute simultaneity. i.e.
as rich as the Newtonian spacetime. which can also be obtained as the result of
enriching the Galilean spacetime by a standard of absolute rest. 3
57
H. Eilstein (ed.), A Collection of Polish Works on Philosophical Problems
ofTime and Spacetime, 57-72.
© 2002 Kluwer Academic Publishers.
58 JAN CZERNIAWSKI
Let us remember, however, that STR is only a local approximation to the general
theory of relativity (GTR). Thus, it is purposeful to ask whether that more adequate
theory does enforce the arguments against the reality of time flow or, on the
contrary, weakens them. The strength of the argument from spacetime symmetry
essentially depends on the answer to the question how rich is the structure ascribed
to spacetime by GTR. From the mere general covariance of Einstein equations does
not follow much, if one remembers the distinction between the covariance of
equations of a theory and the theory's symmetry.4 For, although the group of general
covariance is far richer than the Poincare group, i.e. than the covariance group of the
equations of STR and of any theory consistent with it, each of such theories can be
expressed in a generally covariant form.
term relation of simultaneity. Since apriori it is possible that its results in different
frarnes diverge, in fact every such procedure defines not so much one relation, but
rather a dass of relations, defined in the same way in every frarne. In other words, it
defines not so much a dyadic relation between events, but rather some triadic
relation, the frarne of reference being the third term. 8
Apriori two ways are thinkable, in which operationally defined simultaneity
might acquire absoluteness, here understood as the contrary to relativeness. First, a
procedure might exist that would select some proper subdass of the dass of inertial
frarnes, in which simultaneity defining procedure would give the sarne results. (It is
natural to expect that it would be frarnes at rest relative to each other). This way is
exduded by the principle of relativity. Second, results of some procedure of this
kind might be consistent in the whole dass of inertial frarnes, Le. invariant. But
actually in the spacetime of STR, unlike to the spacetime of non-relativistic
mechanics, no such procedure is possible. Consequently, in the spacetime of STR
the physical relation of simultaneity is relative, whereas the absolute simultaneity, if
it exists, must be non-physical.
As a good exarnple of absolute element of the structure of the special-relativistic
spacetime one might choose spacetime interval. Its operational meaning can be
defined by a certain combination of measurements of the spatial distance and time
interval between two events. Of course, both measurements must be performed in
some frarne of reference, preferably in an inertial frarne, in which the above-
mentioned combination is the simplest. Thus, strictly speaking, this procedure
defines only the interval in a given reference frame. Nevertheless, it turns out that its
results coincide in all inertial frames and one might abstract from the frame in which
it was performed. As a result, spacetime interval acquires an absolute sense in the
second of the above ways, as a relativistically invariant object.
In STR the interval between two events coincides with the 'length', in the sense
of spacetime metric, of the straight line that connects the events. In aspacetime
admitted by GTR straight lines, in general, may not exist. However, it is always
a four-dimensional pseudo-Riemannian space and thus, like the spacetime of STR,
it is a pseudo-metric space. Consequently, it contains geodesics, special cases of
which are straight lines in the special-relativistic spacetime.
The geodesic lines in a general-relativistic spacetime divide into timelike,
spacelike and null ones. An arbitrary line in such aspacetime does not have to
belong to any of these three types. However, a geodesic is distinguished arnong alI
lines of the same type connecting two points by the circumstance that its 'Iength'
takes the extreme value, e.g. maximum among timelike or minimum among
spacelike lines. Just this feature distinguished straight lines in the spacetime of STR.
Thus, the 'Iength' of a segment of a geodesic connecting two events is the most
natural generalization of the spacetime interval between them, provided they can be
connected by a geodesic and that this generalization is unequivocal.
It is easy to see that the interval thus generalized does not have to be defined for
alI pairs of events. But if a given pair is connected by an unique timelike geodesic,
the appropriate interval acquires a simple operational meaning. For the appropriately
chosen units, it is equal to the time interval measured by a dock freely moving in
the given gravitational field in such a way that it coincides (with the required
60 JAN CZERNIAWSKI
accuracy) in succession with both events. In other words: it is equal to the difference
of readings of a clock, whose worldline approximately coincides with the
appropriate geodesic.
In principle, one can attach a reference frame to any smooth timelike line. It is
reasonable to assurne that every observer operating in any frame is endowed with
rigid rods and docks, and that he is capable of identifying and describing any
physical phenomenon within the range of his observation. Having at his disposal
appropriate knowledge and technical means, he can also produce many phenomena.
In particular, he can set one of his clocks into the motion that is needed for defining
the spacetime interval between two given events, at least if it does not require
endowing it with extremely high velocity.
But if it has been done, the dock 'emancipates', so to speak. No matter who has
set it into motion, it measures the interval between those events. The result of the
above procedure is independent of reference frarne, i.e. invariant. This means that, at
least for the special case when the events can be connected by a timelike geodesic,
the spacetime interval between them is not relative, but absolute in the sense under
consideration.
The matter is quite similar in the case of events connected by a null geodesic.
The procedure here is even simpler, although not necessarily easier to perform.
It is enough to send a light signal that will coincide in succession with both events.
The possibility of sending such a signal indicates that the events are separated by
a null interval, regardless who decided to send it.
The case of events separated by spacelike intervals is a bit more complicated.
However, remember that the spacetime of STR is not only a special case
of spacetime of GTR, but also its local approximation in any non-singular point.
In every spacetime region without singularity and small enough for neglecting the
gravitational geometry curvature, in all frarnes attached to spacelike geodesics
(i.e. in the so-called loeal inertial frames) with a certain accuracy the sarne
procedures are aPflicable that enable us to define spacetime intervals in the inertial
frames of STR. Thus, their results are invariant, though not strictly, but
approximately, with the accuracy increasing with the smallness of the considered
spacetime areas.
The possibility of operationally defining the 'distance' between any two events
in a restricted spacetime region means that in such a region also the spacetime
metric has some derivative operational sense. Moreover, regarding what it is
derivative to, it is invariant. Would it mean, thus, that the metric is, after alI, an
absolute element of the structure of general-relativistic spacetime?
Such concIusion would be premature. If only such elements that do not change
under interactions can be regarded as absolute, then certainly the metric in GTR is
not absolute. On the other hand, it follows from the above considerations that the
metric is not relative either. Introducing a new term and distinguishing non-relative
objects from absolute ones seems to be a reasonable way out.
FLOWOFTIME 61
Such program, no matter whether in its less or more consistent version, l;ollides with
the impossibility of measuring the velocity of inertial frame relative to the ether -
the impossibility that results from the principle of relativity. Thus, one must
reconcile oneself with the non-physical, in the framework of STR, character of the
'true' simultaneity and restrict oneself, while doing physics, to the physical
simultaneity defined by uncorrected procedures, notwithstanding its unpleasant
feature of frame-dependence, Le. its relativity, standardly understood. Otherwise,
one would have to assume its relativity understood non-standardly, as dependence
on arbitrary assumption deciding which inertial frame rests relative to the ether.
However, this would mean re-introducing the ether wind into physics, which is
incompatible with the very idea of STR.
satisfying Einstein equations can be legitimately called spacetime and the tensor
field that formally plays the role of the energy-momentum tensor in those equations
can be considered the spacetime representation of matter distribution in a possible
world. Note that this requirement is equivalent to the sometimes set condition of
stahle causalityl9 of spacetime. Thus, all arguments set forth in favor of one of these
conditions speak for the other, as weIl.
In the case of transformations from the Poincare group, such resolution does not lead
to any problems. Any phenomenon in the new spacetime looks in a given inertial
frame exactly the same way as some phenomenon in the old spacetime in another
inertial frame. Moreover, transformations from this group are isometries and thus
the metric also is the same as before the transformation.
This sirnilarity has inclined some authors to a hasty denial of the difference
between the two above-mentioned interpretations of transformations. 21 However,
one should not forget that even from the relationist point of view, which usually
motivates such a move, the active interpretation must mean that the relations
between the observer in question and the rest of the world have changed. Now they
are the same as between some other possible ob server and the rest of the world
before the transformation. Consequently, for justifying such a denial aversion
of relationism would be necessary radical enough for substantiating the exdusion of
identity of an observer after his changing reference frame in physically the same
world.
Problems arise in the case of transformations from outside the Poincare group.
The active interpretation of a transformation means here moving to a world in which
all physical phenomena have been distorted by an universal force. As a result,
physical measuring rods and docks are no longer reliable means for defining the
spacetime metric, but the results of measurements performed with their help must be
suitably corrected.
The metric defined by corrected measurements expresses itself standardly in the
chosen coordinates, thus satisfying Einstein equations. But the latter are also
satisfied by the quantity defined standardly by non-corrected measurements, the
quantity not yet regarded as metric. In the new spacetime physical phenomena
behave anomally. In particular, in spite of lack of sources of a gravitational field, the
worldlines of free particles are not geodesics; they only would be geodesics if the
metric were defined by non-corrected measurements.
Moreover, practically rigid measuring rods do not now have to retain their
lengths after changing their position and the moment of time. Sirnilarly, physical
docks do not have to go equally, independently of the place and time. All these
anomalies are correlated with the distribution of that quantity not yet regarded as the
metric, which thus may be treated as a quantity expressing the universal force. What
is more, this correlation is of such a kind that a freely moving observer is unable to
decide empirically whether he is in the world before or after the transformation.
Then, the once considered operation of geometrization of the universal force
suggests itself. In its consequence, the metric defined by the non-corrected
measurements is considered physical. On such an assumption, the spacetime after
the transformations begins to look exactly the same as before. Thus, one might be
tempted to identify both spacetimes. 22 Yet, this would be amistake opposite to the
one committed by the advocates of the thesis of relativity of metric as an alleged
consequence of GTR. For such identification would mean that, in fact, no actively
understood transformation took place, but only some transformation understood
passively, as a mere change of parametrization of the spacetime.
66 JAN CZERNIAWSKI
at any such moment it has the propensity, expressed by its momentum, to change its
place - the propensity which for the same reason is not had by an arrow at rest.
Yet, one may ask why this propensity is all the time approximately the same. Weil,
just because the arrow is constantly the same! On the other hand, any changes of this
propensity result from well-defined interactions of the arrow with the air and the
gravitational field of the Earth. Were it not so, no logical reason would be able to
rule out the possibility that the thing which was the arrow at the beginning of
its flight would behave in some completely crazy way.
Now we see the reason why one does not have to take seriously the cosmological
models which reveal causeless irregularities inside the 'hole'. From the reistic
perspective the world can be understood as a dynamic system of bodies and fields,
the gravitational field incIuding. The spacetime metric is nothing else but expression
of the propensities in their behavior wh ich result from some particular state of the
gravitational field. Thus, if it manifested some regularity outside the 'hole', then
such regularity would result from certain dynamics of the system and there is no
reason why it should disappear in the arbitrarily separated area of the 'hole'.
Jan Czemiawski
Institute of Philosophy, Jagiellonian University,
Grodzka 52, 31-044 Cracow, Poland
E-mail: uzczemi@cyf-kr.edu.pl
FLOWOFTIME 71
NOTES
I Cf. Rietdijk (1966).
2 For the meaning ofthis term see W6jcicki (1979), p. 28.
3 Earman (1989), p. 33.
4 Anderson (1967), p. 73.
5 Ibid.
6 Cf. Landau, Lifschitz (1959), pp. 281-282.
7 Cf. Selleri (1996) and Selleri (1998).
8 Denbigh (1981), p. 44.
9 Misneretal. (1973), § 16.4.
10 Cf. Anderson (1967), pp. 84-88; Earman (1989), p. 29.
15 This is why the conception in question is often called conventionalism - cf. ibid., pp. 35-36.
16 Ibid., p. 256.
17 Cf. Tangherlini (1961), pp. 8-9; Selleri (1996) and Selleri (1998).
18 Gödel (1949).
19 Cf. Wald (1984), p. 198.
20 Earman, Norton (1987).
21 Cf. e.g. Wald (1984), p. 439.
22 Cf. Earman, Norton (1987), p. 522.
23 Ibid., p. 523.
24 Raine (1981).
REFERENCES
Anderson, J.L. (1964). Riemannian Geometry. In: H. Chiu, W.F. Hoffman (eds.), Gravitation and
Relativity. New York: Benjamin, pp. 17-39.
Anderson, J.L. (1%7). Principles olRelativity Physics. New York: Academic Press.
Butterfieid, J. (1984). Seeing the Present. Mind93, 161-176.
Czemiawski, J. (1998a). On What There Is Not - a Vindication of Reism. In: K. Kijania-Placek,
J. Wolenski (eds.), The Lvov-Warsaw School and Contemporary Philosophy. Dordrecht: Kluwer
Academic Publishers, pp. 313-317.
Czemiawski, J. (1998b). Über Stäbe, Uhren und Relativität. In: Selleri et al. (1998), pp. 111-159.
Denbigh, K. (1981). Three Concepts olTime. Berlin: Springer Verlag.
Earman, J. (1989). World Enough and Spacetime. Cambridge (Mass.): MIT Press.
Earman, J., J. Norton (1987). What Price Spacetime SubstantivaIism? The Hole Story. British Journal
lor the Philosophy 01 Science 38, 515-525.
Gödel, K. (1949). An Example of a New Type Solution of Einstein's Field Equations of Gravitation.
Review 01 Modern Physics 21, 447-450.
Kotarbinski, T. (1966). The Development Stages of Concretism. In: Gnosiology. The Scientific Approach
to the Theory 01 Knowledge. Oxford: Pergamon Press, pp. 429-437.
Landau, L.D., E.M. Lifshitz (1959). The Classical Theory 01 Fields. London: Pergamon Press.
72 JAN CZERNIAWSKI
Maxwell, N. (1985). Are Probabilism and Special Relativity Incompatible? Philosophy 01 Scienee
52, 23-43.
Misner, C.W., K. Thorne and J.A. Wheeler (1973). Gravitation. San Francisco: Freeman.
Popper, K.R. (1991). A World of Propensities: Two New Views of Causality. In: Advanees in Scientifie
Philosophy. Amsterdam: Rodopi.
Raine, DJ. (1981). Mach's Principle and Spacetime Structure. Reports on Progress in Physies 44,
1151-1195.
Reichenbach, H. (1958). The Philosophy 01 Spaee and Time. New York: Dover Publications.
Rietdijk, C.W. (1966). A Rigorous Proof of Determinism Derived from the Special Theory of Relativity.
Philosophy 01 Scienee 33, 341-344.
Rindler, W. (1977). Essential Relativity. New York: Springer Verlag.
Selleri, F. (1996). Noninvariant One-Way Velocity of Light. Foundations 01 Physies 26,641-664.
Selleri, F. (1998). Nichtinvarianz der Ein-Weg-Lichtgeschwindigkeit. In: Selleri et al. (1998), pp. 75-110.
Selleri, F., J. Brandes, J. Czerniawski, U. Hoyer and K. Wohlrabe (1998). Die Einstein 'sehe und
Lorentzianisehe Interpretation der speziellen und allgemeinen Relativitätstheorie. Karlsbad: VRI.
St. Augustine (1961). Conlessions. Baltimore: Penguin Books.
Tangherlini, F.R. (1961). An Introduction to the General Theory of Relativity. Nuovo Cimento Suppl.
20, 1-86.
Wald, R.M. (1984). General Relativity. Chicago: The Univ. ofChicago Press.
W6jcicki, R. (1979). Topies in Formal Methodology. Dordrecht: Reidel.
TOMASZ PLACEK*
Abstract. In this paper I analyze this variety of transient time theory that relies on the notion of open
future. I present algebraic models of phenomena with transient time, understood as above. The models are
then linked to relativistic spacetimes. I finally address some interpretation al issues and defend the theory
ofbranching time against David Lewis' objections.
Keywords: transient time, branching spacetime.
1. INTRODUCTION
Transientism is a doctrine that the central aspect of time is, or derives from,
the objective transition of the future into the (momentary) present and then the past,
with future events becoming actualized in some 'now', and then passing into
the past. Following the popular terminology of (McTaggart, 1908), such
a succession of events is referred to as an A-series and contrasted with the earlier-
later relation, whose relata form a B-series. The description of events by means of
a B-series, if once valid, is always valid: if one event is earlier than some other
event, it was always so and it will ever be so. In contrast, the description of events
by an A-series must change: an event that is now in the future, will be once at
present, and later on in the past. The appeal of transientism comes from the fact that
a mere B-series hardly makes justice to our intuition of the passage of time.
Aigebraically speaking, the relation earlier-later is indistinguishable from the I
below-above relation, but clearly a vertical (and infinitely thin) stick, with its points
being ordered by the below-above relation is not an adequate model of finite time.
In contrast to this popular, as I take it, appeal of transientism, the doctrine fares
badly in physical sciences. As far as I know, neither the particle physics, nor
gravitational theories, nor familiar cosmological models offer a clear perspective for
accommodating transientism, or, as it in essence boils to the same thing, becoming.
At this point it is perhaps worth observing that the task of accommodating
transientism goes far beyond the introduction of the arrow of time, as the latter is
concerned with finding a physical basis of the earlier I later distrinction. What is
then the task of accommodating transientism? To get clear about it, and leave a
rather misty field, where physical sciences confront our temporal intuitions, we need
to clarify transientism considerably, preferably by finding a class of models of
temporal becoming that are both mathematically precise and intuitively correct. It is
my hope that the models of stochastic outcomes in branching spacetime (SOBST),
which I present in Sections 2 and 3 satisfy these desiderata to a reasonable extent.
I also believe that these models shed some light on what accommodating for
transientism involves on the part of physical theories. But, since the extant physics
does not support the transientist' s view of time, must the transientists bet on future
73
H. Eilstein (ed.), A Collection of Polish Works on Philosophical Problems
ofTime and Spacetime, 73-92.
© 2002 Kluwer Academic Publishers.
74 TOMASZ PLACEK
physics refonning itself to the effect of accounting for transientism? Or, may they
maintain that their favorite doctrine is independent of future findings of physics?
Helena Eilstein's powerful claim that '[i]f there is Becoming, but physicists cannot
know it, nobody can' (Eilstein, 2(02) is a clear voice for the first option. However,
I advise to take as adefault option the independence of transientism of physical
findings. I take transientism to be a modal concept that builds upon the notion of
open future. Accordingly, those who tie the fate of transientism with the progress of
natural sciences believe that at some stage these sciences will arbitrate between
competing modal claims. As I see the matter, our evidence from extant physics
supports the opposite belief. Take for instance the modal claim that natural
phenomena are indetenninistic together with the popular view that this
indeterminism is intimated by quantum physics. The view is half true about Hilbert-
space quantum mechanics, as its measurement algorithm usually yields a set of
possible results of a given measurement, without any indication which of these will
occur on a given occasion. Yet, the other half of the truth is that temporal evolution
of quantum states, as described by the Schrödinger equation, is fully deterministic.
The conflict between the measurement algorithm and the smooth temporal evolution
of states is known as the measurement problem of quantum mechanics and seen as a
scandal in the foundations of physics. But the situation is even worse, as there is a
perfectly detenninistic Bohmian quantum theory whose predictive power is exactly
like that of Hilbert-space quantum mechanics. Thus, witnessing that quantum
theories yield so disparate verdicts about the truth of determinism, it is doubtful
whether the more subtle view of transientism could ever be validated or refuted by
physics. Moreover, if the theory of transient time is what I take it to be, namely a
modal doctrine, it is rather collaborated efforts of logicians and logically minded
philosophers that might intimate whether or not it is true. Be that as it may, to obtain
clarity about the concept of transient time is necessary for anyone investigating if
this view is true.
As I already hinted, aversion of transientism that I am interested in draws on the
concept of open future. In this vision the distinction between past and future consists
in an asymmetry of the two with respect to possibilities. A future event has a few
possible outcomes, but as soon as it becomes past, only one of its outcomes is
actualized, the remaining outcomes being no longer possible. The present is where
this transition from many possibilities to a single actuality takes place. Thus, to
properly model this concept of open future, we need to produce a structure with
events and their outcomes, and then introduce a means of 'deleting' the non-
actualized outcomes. One might want to express this vision in tenns of possible
histories, yet the description in tenns of events and their outcomes is more in line
with SOBST models, to which I soon turn. The SOBST framework is a development
of Belnap's branching space-time and his outcomes in branching time (Belnap,
1992; Belnap, 1995).
The paper is organized as folIows. Seetion 2 briefly sketches the purely algebraic
models of stochastic outcomes in branching spacetime, as developed in (Kowalski
and Placek, 1999; Placek, 2000; Müller and Placek, forthcoming). In the next
section the models are related to geometrical notions of spacetime physics. Finally,
BRANCHING FOR A TRANSIENT TIME 75
Section 4 defends this version of branching time theory against some popular
objections and illuminates a number of ontological assumptions behind SOBST.
x y z
Figure 1: Upward and downward forks; the lines represent the ordering relation<.
THEOREM 1
The family FE of outcomes of E c W forms a complete and atomic Boolean algebra
where nis the set-theoretical intersection. and. aUE b =( aU b 19lo g the unit
element of the algebra is the set W. and the zero element of the algebra is the empty
set.
THEOREM 2
For every complete atomic Boolean algebra B there is a world Wand a set E c W
such that the algebra BE of outcomes of E is isomorphie to B.
An event is defined as a subset of a history that is bounded from above. 1
We can also introduce spacelike events E and F by saying that E and F are
spacelike iff there is a history that contains them both and no part of one event is
preceded by any part of the other event.
To comment on our concept of an outcome, note a difference between this
(technical) notion and an ordinary notion of an outcome or a result. A SOBST
outcome is upward closed, and hence it continues as long as histories it involves
continue. To the contrary, an outcome of a measurement, say a scintillation or
a click of adetector, is a small well-Iocalized chunk of a history. Thus, the two
78 TOMASZ PLACEK
notions cannot be identified. Yet, it is very reasonable to require that atomic SOBST
outcomes of an event are in a one-to-one correspondence with results of that event.
(An atomic outcome of E is any non-empty outcome of E that does not contain any
other outcome of E). This stipulation will perrnit us to link experimentally
established probabilities of measurement results with probabilities of SOBST
outcomes.
To introduce probabilities, it suffices to equip each Boolean algebra BE associated
with the family of outcomes of an event E with a normalized, countably additive
measure PE: BE ~ [0,1], i.e., a measure satisfying
j ) j ).
j;J j;J
The introduction of probabilities to the model W has the effect of making the
resulting model more fine-grained. Thus, a SOBST model is a quadruple(W,S,e),
with Wa non-empty set, =s;- a partial ordering on W, c - a set {(E,PE)} of pairs,
each pair consisting of an event in Wand a probability measure assigned to it.
But how should we interpret probabilities of outcomes of events? Tentatively,
we think of these probabilities as objective measures of possibility, that is, we
interpret PE( e ) = x as saying that the chance that the outcome e of event E occurs is
x. I said above 'tentatively', because it can happen that details of investigated
phenomena tell against this interpretation. As presented in (Placek, 2000a), an initial
model with non-trivial probabilities sometimes can be extended into a larger model
that contains trivial probabilities only. Further, models with trivial probabilities can
be made completely deterrninistic by erasing all outcomes with probability zero.
In case such a re-description is possible, the SOBST framework alone cannot decide
which model is correct. But, if we have additional grounds for preferring a
deterministic model in a given case, then the non-trivial probabilities of the initial
model must be taken as epistemic probabilities, that is, as ones representing our
ignorance of factors that would provide for a more fine-grained description of
deterministic phenomena.
Note that at this stage we have a (perhaps huge) number of probability spaces,
each probability space being tied to the family of outcomes of some event in W.
Within a given probability space (BE,PE) we can also define conditional
probabilities: the probability that outcome x of E happens, given that outcome y of E
happened. But this approach allows only for the assignment of probabilities to
BRANCHING FOR A TRANSIENT TIME 79
FACT Ilf e andf are outcomes of E and F, respectively, then enf is an outcome
of Eu F . And, if a is an outcome of Eu F , then there are outcomes e of E and f of
F such that a =enf .
Suppose now E and F are events with outcomes ef, ... ,eN and/J, ... JM, respectively.
Let us also assume that every history containing E contains F, but not necessarily
vice versa. This obviously guarantees that E u F is an event and that the same
histories contain E and Eu F. Further, by the facts above, any intersection
e j n f j is an outcome of Eu Fand (given the assumption above) for any non-
empty outcome ej of E there is at least one outcome jj of F such that ej n fj '# 12) •
Combining these observations with the facts above, we impose this requirement on
SOBST models:
REQUIREMENT
If any history that contains E contains F as well and E and F are events with
outcomes ef, ... ,eN andfl, .. .JM, respectively, then
probabilities are concerned, F can be ignored. (3) Perhaps the most interesting
application of REQUIREMENT is to spacelike events E and F, e.g. to a correlation
experiment with two measurements performed in spacelike separated regions.
The 'single' probabilities are measured in each wing and described by IlE( ej )
and IlE( fi)- to measure them one observes a counter set in the corresponding
wing of the setup. There are also probabilities of coincidences, represented by
IlEUF( ej n I} ). One measures them at a coincidence gate, where the data from the
two wings are collected. Since this is a correlation experiment, E is measured
whenever F is measured, which makes the premise of REQUIREMENT satisfied.
Hence we have that the probability of a result produced by E should be the same as
the sum of probabilities of those coincidences e n I j (produced
j by the
measurement Eu F ) that involve the result in question. 2
Although REQUIREMENT is reminiscent of what is known in probability
ca1culus as marginal property, we consider it a semi-empirical claim that is capable
of refutation. Yet, there is every good reason to accept it as applied to observable
probabilities - no experiment has given the slightest evidence against it. However,
in some applications in (Placek, 2000a) REQUIREMENT refers to unobservable
probabilities, where one might be more sceptical as to its validity. The sceptical
attitude will probably be reinforced by noting that this condition taken together with
some further premises permits the derivation of the experimentally violated Bell's
inequalities. It is, however, extremely difficult for the sceptic to explain how, in
some hidden layers of reality, REQUIREMENT could possibly fail, given the
premise that no his tory containing E escapes F.
Figure 2: Spacetimes split at a single choice point x (on the left). and at two choice points x
and y (on the right). The shared regions are shadowed and solid lines indicate the surface 0/
divergence.
To say this more precisely, let us focus upon two Minkowski spacetimes, say,
0" and T}, whose points are ordered, respectively, by ~O" and ~1J' and that split at
the choice points forrning a non-empty set CO"1J ' These spacetimes should be thought
of as two copies of the Minkowski spacetime, with points of one spacetime being
related to points of the other by a 'counterpart relation' RO"1J • The expression xRO"1JY
is to mean that point xE 0") and point y E 0"2 are counterparts. Now, for c to be
a choice point between 0" and T}, it must be that CE 0" and cE T}. Recall also that
choice points of two spacetimes must be space1ike separated. Thus, taking the notion
of choice points for primitive, the common segment of 0" and T} is produced by this
requirement:
ForxEO"and yET}.x=yiff xRO"1JY and neither x nor y is in the
forward light cone of any choice point cE CO"1J '
= stands for identity, so the above condition says that expressions 'x'and 'y' denote
the same point that belongs to the two spacetimes. We further require that no point is
shared by 0" and T} if CO"1J is empty. Given that a common segment of two
spacetimes is as above, their surface of divergence must be as folIows :
BRANCHING FOR A TRANSIENT TIME 83
For Minkowski spacetimes G and 17, and the set Cer" of their choice
points, z belongs to the surface of divergence D(G.17) iff z lies on the
forward light cone of some cE Cer" and does not lie within the
forward light cone of any CE Cer".
Note that a surface of divergence is constructed out of light cones. This means
that relations 'lying on (below or above) a surface of divergence' are Lorentz
invariant. There is much to be done to extend these observations to a general
definition, which should make it dear what the result of pasting a farnily of
Minkowski spacetimes iso In this object we will have many pairs of Minkowski
spacetimes and thus - many sets of choice points for pairs of spacetimes. We will
thus need to decide how these sets of choice points relate. Also, since a history is not
merely aspacetime, to represent a history we had better make a complete partition
of aspacetime into regions, and then assign astate to each region so produced.
Finally, we will need to introduce events and their outcomes to this geometrical
framework in such a way that oUf main algebraic results (Theorems 1 and 2) be true
about them. Yet, there is no need to deal with these tasks here, since they have only
a marginal significance for the present purpose of analyzing transientism. For this
purpose, however, two consequences of the present concept of surface of divergence
are of utmost importance. First, lying on (or below, or above) a surface of
divergence is Lorentz invariant. This means that relations like 'something is possible
at a point x' hold or not, independently of the frame of reference.
Second, for any two surfaces of divergence D( G, 17 ) and D( G, r ), either one is
below the other, or they intersect. This fact suggest adefinition of a transient now,
or of the set of points co-present with a given one: a region co-present with a given
point x is the region limited by the first surface of divergence above x and the first
surface of divergence below X.
As time passes, events become actualized, and given that an event occurs, exactly
one atomic outcome of it is actualized, while all the remaining outcomes become ex-
tinct. This leads to a process that is opposite of branching, namely the attrition of
histories. What are the histories subject to attrition?
ATTRITION
r
Given that an event E from history G occurs, any history for which the surface of
r)
divergence D( G, is in the past of E, is erased.
It follows that histories undergo attrition along the same surfaces as they diverge,
namely, along surfaces of divergence.
4. QUESTIONS OF INTERPRETATION
I sketched above a framework for transient time whose two essential elements are
outcomes of events and attrition of non-actualized outcomes, from which attrition of
histories folIows. Yet, I touched interpretational issues only in passing.
84 TOMASZ PLACEK
In this section laddress them by replying to objections that are raised to SOBST
models as weIl as to branching in general.4
(Q I) A SOBST model usuaIly has a number of histories. Does this mean that aIl
these histories are believed to be real?
(RI) First, jOBST offers a technique of deciding whether a given stochastic
experiment aIlows a deterministic description. If it does, then the experiment has a
deterministic model, Le., one that contains a single history only. Thus, the question
of reality of histories is troublesome only if we have a model that cannot be
extended to a deterministic one. In such cases we may indeed say that SOBST is
committed to many histories. However, since the other side of proliferating
branching histories is the attrition of those branches of histories that have not been
actualized, it is still better to say that SOBST is committed to the vision of an open
future, according to which an event, though perhaps not every event, has more than
one alternative outcomes. In other words, there may be many alternative future
possibilities of an event, and this we may express by saying that there are many
alternative possible futures. Yet, as time passes, at most one atornic outcome of an
event becomes actualized, which agrees with our intuition that the event has exactly
one real future, and that we have exactly one real future. Yet the question of why
from among our many possible futures this one rather than some other one becomes
real, is not something that can be answered, at least if indeterrninism is true.
(Q2) Openness of the future is temporal, that is, an outcome of an event that
rnight have been actualized once this event occurred, but did not, is no longer
possible. How is this represented in SOBST?
(R2) The degree of openness of the future depends on time, that is, once a given
outcome of some event is actualized, the actualization of any of its alternative
outcomes is no more possible. This idea is captured in the concepts of branching of
histories and of the attrition of those branches that have not been actualized.
To represent an open future, histories are assumed to branch along surfaces of
divergence. As time passes, those branches that are not actualized are erased.
(Q3) How does attrition of branches propagate in spacetime, or, in other words,
what is the surface along which attrition occurs?
(R3) It is exactly the same surface along which histories branch and along which
aIl but one branch are erased. In the special case of a single choice point of
Minkowski spacetimes, histories branch and undergo attrition along the future light
cone of this point, but in general surfaces of divergence and attrition are more
complicated. Importantly, surfaces of divergence of a given history u with other
histories succeed one another, and this succession is independent of the choice of
frame of reference in u. Note also that there is no twilight zone between a still
possible history and an already impossible history: attrition is razor sharp.
(Q4) How does SOBST represent the natural distinction between an actualized
event (which is in the past or at present) and an event that is in the future but is
bound to happen?
(R4) The distinction builds upon a difference between a single (one out of many
formerly possible) actualized outcome of an actualized event and a single possible
outcome of a yet non-actualized event. In the latter case, if the event in question
occurs, its single possible outcome invariably actualizes.
BRANCHING FOR A TRANSIENT TIME 85
(Q5) If events with single atOInic outcomes are allowed, and the future is
differentiated from the past and the present by open possibilities, how can there be
the future in those regions of W that yield only single atomic outcomes of events
involved?
(R5) Our modal structure W has a distinguished stratification by a sequence of
surfaces of divergence, these in turn being produced by a set of choice points for
spacetimes involved. Given that it is known how to define, by means of the attrition,
the last and the next surface of divergence for any given point x in aspacetime er ,
it is natural to require that the present to which x belongs is the region of W that
is between the last and the next surfaces of divergence in er for this point x.
In general, this will make the present temporarily extended. However, since the
surface of divergence is produced by a set of spacelike choice points, the believer in
ubiquitous chanciness can argue that for any pointlike event with a single atomic
outcome there is a larger event comprising the said event and some spacelike
separated point such that this larger event has more than a single atomic outcome.
By requiring further that the distance between the said event and the added point be
arbitrarily smalI, we will get the arbitrarily small temporal extension of any present.
The other option, and one that I recommend, is to concede that the present can be
temporarily extended. If there are deterministic regions, the extended present is to be
expected. In the limiting case, if W has no choice points, we get the permanentist
Uni verse, with any event being at present, i.e., actual, and there being no objective
distinction between past, present, and future.
(Q6) Does SOBST define the arrow of time?
(R6) No, SOBST assurnes the arrow of time. Indeterministic SOBST models are
asymmetric with respect to time: the arrow of time points to where the open
future iso Yet, without indicating, by means of precedence relations, the direction of
the flow of time, the notion of future and open future cannot be formulated in
SOBST.
(Q7) Does SOBST offer, or does it support, any interpretation of quantum
mechanics? It appears similar to Everett's interpretation.
(R7) The similarity is rather illusory. Everett's interpretation is usually taken as
positing a number of equally real alternative scenarios, one of which is distinguished
merely by our being in it. For instance, the measurement of spin polarization on
a spin-.i particle produces two equally real scenarios, one with the result 'spin up'
2
and the other with the result 'spin down', and one of this scenarios appears real
because we are in it. In SOBST we may similarly think of future possibilities, but
once the measurement is over, there is exactly one scenario (one branch), the other
having become extinct.
Turning to the question of a prospective SOBST interpretation of quantum
mechanics: at the moment there is none, and it is not clear if SOBST can be
successfully employed to this end. Any interpretation of quantum mechanics must
face the measurement problem, which is, in essence, a conflict between the
continuous and deterministic evolution governed by the Schrödinger equation on the
one hand, and chancy occurrences of measurement results, as described by the
measurement algorithm, on the other. SOBST can easily represent the measurement
86 TOMASZ PLACEK
such a way that the probability of the atomic outcome of A that contains B is greater
than the probability of the remaining (non-atomic) outcome of A. And what is
achieved by conditionalization in other frameworks, is achieved in SOBST by
moving along a given history from one probability measure to another.
(QlO) SOBST uses a physical notion of state, and moreover, it uses it incorrectly
by assigning states either to spatiotemporal points, or to spatiotemporal regions.
Physics assigns states to systems.
(RlO) The algebraic approach, indeed, reHes on the concept of spatiotemporal
point taken together with what is in it. In essence, this is the point-concrete
particular, and it is fairly weil known from ontology. A pointlike particle of classical
mechanics offers an analogue in theoretical physics. In our geometrical approach
states are assigned to regions of spacetime, and although perhaps it is not the
standard, the concept can be found in the physics literature, see e.g., (Penrose and
Percival, 1962).
(Qll) The effect of pasting together segments of branching space-times is that
the result is no longer a topological manifold, yet this is the very basis required by
general relativity.
(Rll) Speaking quite generally, SOBST is not a physical theory, and hence it
does not compete with any physical theories, quantum gravitation included. It is
designed to make precise the vision of open future and to analyze certain arguments
about physical experiments, especially those that invoke modal, spatiotemporal, and
probabilistic notions. Modality is hardly in the repertoire of extant physics, and
explaining the workings of physical systems is hardly a task for SOBST. Thus,
differences between the mathematics required by physics and that required by
SOBST should be expected.
On the particular question of manifolds, however, I believe that the vision of
branching spacetimes is correct and moreover necessary to make sense of chancy
phenomena. Aspacetime is a scene of physical processes and in this sense it
describes either the real world or a physically possible world. But to represent
alternative possibilities one needs a collection of spacetimes, which, if pasted, do not
form a topological manifold. And since at present general relativity requires
topological manifolds, I tend to say: the worst for present physics. I believe that
some future theory of quantum gravity will overcome this problem, and, as a matter
of fact, there is already some work going in this direction. 5
(QI2) Branching is an alternative to the much more popular Lewisian view of
possible worlds that diverge but do not branch. What is exactly the difference?
(RI2) Some branching histories do overlap, and then they have a common initial
segment. On the other hand, Lewis' divergent worlds are analyzed in terms of
duplication. Duplicates are separate, that is, non-overlapping, objects that agree with
respect to natural properties and relations. Two worlds are then said to diverge iff
they are not duplicates but an initial segment of one world and an initial segment
of the other are duplicates (Lewis, 1983, p. 359).
(Q13) It is commonly believed that Lewis showed that branching is problematic
in a way that diverging worlds are not. Does SOBST have any answer to his
arguments?
88 TOMASZ PLACEK
(R13) Lewis' objections are related to the task of representing the semantics
of modal discourse in models with branching, and we have not built a SOBST
semantics for modal (and temporal) discourse yet. Thus, whatever I say on this topic
is very tentative. Lewis' first argument against branching concerns individuation,
so I need to say what individuals are in the SOBST framework. SOBST models do
not distinguish between natural and artificially carved individuals, and hence
an individual is just a part of a history, that is, an event. An individual is thus
contained in many histories, in general. To illustrate, Gottlob Frege, in his
spatiotemporal entirety and as he really was, belongs to many histories that have
branched or will branch after his death, viz. histories that have not yet been erased.
However, only one of these histories will ultimately survive the process of the
attrition of branches-this is the real history. Further, to accommodate for the
intuition that Frege could have been different, we postulate that some possible
histories split in Frege's lifetime, and that each of these histories contained some
initial segment of the real Frege. As Frege is long deceased, all but one of the
branches that split 'in' Frege were erased. But, significantly, the histories that split
in real Frege's lifetime contained, strictly speaking, different individuals (events),
their common feature being that they shared some initial segment of the real Frege.
This should answer the first argument of (Lewis, 1986, p. 199), the hero of which,
though, is Hubert Humphrey and not Gottlob Frege:
He eould have had six fingers on his left hand. There is some other world Ihat so
represents hirn. We are supposing now that representation de re works by trans-world
idenlity. So, Humphrey, who is apart of this worid and here has five fingers on the left
hand, is also a part of some other worid and there has six fingers on his left hand. Qua
part of this worid he has five fingers, qua part of that world he has six. He himself [ ... 1
has five fingers on the left hand, and has not five, but six. How ean Ihis be?
stage S, that is shared by at least two histories,which split immediately above this
event. The two histories contain different continuations of Humphrey in stage S, one
continuation with five fingers on the left hand, and the other with six fingers on the
left hand. Accordingly, one history has the individual (event) with five fingers on his
left hand in his after-stage S life and the other history has the individual (event) with
six fingers on his left hand in his after-stage S life. Thus, the two contradicting
properties, of having five fingers and having six fingers on the left hand, refer to two
different individuals that share some initial segment, and only one of them is in a
history that has not been erased. Hence, no contradiction ensues. Of course, this is
only a sketch of a full reply, since we need to be precise as to what the continuations
are; otherwise your reading this paper is a continuation of Humphrey at stage S as
weil. Yet, even without elaborating any further on this problem, 1 believe, the
answer suffices to disarm Lewis' objection.
One may nevertheless say that the problem resurfaces once we consider
Humphrey before stage S, say pre-S-Humphrey. It appears that pre-S-Humphrey has
five fingers on his left hand qua being in one history and has six fingers on his left
hand qua being in another history. However, this conclusion follows in a non-tensed
picture of language, where we consider sentences like 'Pre-S-Humphrey, who
finishes on Jan. I, 1913, has five fingers on his left hand on Oct. 1, 1969'. SOBST is
not committed to non-tensed semantics, but, on the contrary, naturally suggests
tensed semantics. So, in accord with our everyday way of speaking, from the fact
that 'Humphrey could have had six fingers on his left hand' is represented by there
once being a continuation of pre-S-Humphrey with six fingers on his left hand,
it does not follow that pre-S-Humphrey had six fingers on his left hand.
Lewis' second worry concerns the demarcation of worlds-see (Lewis, 1983,
p.360):
( ... ) [O]verlap of worlds interferes with the most salient principle of demarcation for
worlds, viz. that two possible individuals are part of the same world iff they are linked
by some chain of extemal relations, e.g. of spatiotemporal relations.
Supposedly the branching theorist is in for trouble with the phrase 'the future
of ... '. Grammatically, it looks like adefinite description, and this standardly
involves two things: there should be exactly one object that is the future of ... (call
this the uniqueness condition), and the property of being the future 0/ ... should
suffice to pick out this single object (dub this the sufficiency condition). In
branching models, there is no problem with uniqueness, since there is exactly one
real history, and hence anything real has exactly one real future. But sufficiency
appears to be a problem: if indeterminismis the case, no matter how much we extend
the phrase 'the future of ... ' and how weIl we understand it, we will not be able to
say what this future iso Moreover, if there is no end to the world, no one will ever be
able to do so. Metaphorically, even God, whose predictive capacities are limited by
indeterminism, cannot do so. Yet, in my view, it is amistake to take 'the future
of .. .' as adescription. (Lewis, 1986, p. 207) considers three ways of understanding
this phrase:
If there are two futures, and both are equally mine with nothing to choose between
them, and one holds a sea fight and the other doesn't, what could it mean for me to say
that the future holds a sea fight? Not a rhetorical question: we have three options. (1) It
is false that the future holds a sea fight; because 'the future' is a denotationless improper
description. (2) ltis true that the future holds a sea fight; because 'the future' denotes
neither of the two partial futures hut rather their disunited sum, which does hold a sea
fight. (3) It is neither true nor false that the future holds a sea fight; because 'the future'
has indeterminate denotation, and we get different truth values on different resolutions
of the indeterminacy.
Then one sense of a person's assertion 'There will be a battle tomorrow' has this
explanation: in an outcome of some specified event containing the act of assertion in
question there is an event of battle that is located in some specific part of this
outcome, and moreover, this event belongs to the history that is not erased.
On this translation the question of whether 'tomorrow' refers to apart
of some single outcome, or to parts of many outcomes is answered by saying that on
its own'tomorrow' does not refer. However, 'tomorrow's battle' as uttered in this
mode of speaking, refers to an event that is (1) some specific part of an outcome of
the event of asserting the sentence and (2) this part of this outcome belongs to the
his tory that is not erased.
To make this analysis precise, we need, however, to first erect a semantics
of tensed discourse on the SOHST framework. 8 Thus, at present we do not have
a watertight answer to Lewis' objections, but merely a sketch of what this answer
will be, once SOHST semantics is implemented. Yet, I believe that even at this
prelirninary stage it can be seen that Lewis fails to show that branching, SOHST
style, is more problematic than diverging possible worlds. And since SOHST is
mathematically more precise, has nice mathematical properties, and is applicable to
experimental data, it is preferable, I believe, to Lewis' framework of diverging
possible worlds.
5. RESULTS
Taking as a starting point the idea that the transient aspect of time builds upon
openness of future, I presented above the framework of stochastic outcomes in
branching spacetime that is intended to serve as a mathematical model of the world
with transient time. Section 2 presented purely algebraic SOHST models.
Algebraically defined histories of these models were then linked in in Section 3 to
relativistic spacetimes. Section 4 answers some objection to SOHST models and
branching in general. In particular, it disposes of Lewis' arguments that branching
faces some unsurmountable difficulties of wh ich its main and more popular riyal,
the doctrine of diverging possible worlds, is free. As I stressed, the vision of
transient time that I developed here is based upon openness of future, and hence,
upon (a version) of indeterrninism. Hut what if, as a matter of fact, indeterrninism is
false? The answer is, I think, that the doctrine of transient time is then false as weil.
If this were so, nothing in the physical reality would correspond to our intuition of
the passage of time as weil as to some temporal forms of our language.
Tomasz Placek
Department of Philosophy,
Jagiellonian University
Grodzka 52,31-044 Cracow, Poland
E-Mail: uzplacek@cyf-kr.edu.pl
92 TOMASZ PLACEK
NOTES
This paper contains some passages from my book Is Nature Deterministic?, (2000), Cracow,
Jagiellonian University Press. It has has been stimulated by discussions I had with Jacek Cachro and
Thomas Müller. I wish also to thank Professor Helena EiIstein for her many careful and extremely
detailed comments and corrections.
1 Note that this definition is in conflict with the physicist's use of the term.
2 Recall the one to one correspondence between results and atomic outcomes.
3 Pedantically, '!Ix. ye Cu .17 {x}and {y} form a spacelike set.
4 Questions below come from my exchange with Storrs McCall and audiences that Iistened to earlier
versions of !his paper. I am especially indebted to participants of the symposium 'Time', Cracow, May
6-8, 2000, the annual meeting of The British Society for the Philosophy of Science in Sheffield, July
8-9, 2000, and the meeting of Gesselschaft für Analytische Philosophie in Bielefeld, Sept. 25-28,
2000.
S Compare Heller 1988, p. 34.
6 The distinction comes from Xu, 1997.
7 Hubert Horatio Humphrey, Vice-President of the Uni ted States under Lyndon Johnson (1965-69). Iived
from 1911 to 1975.
8 I owe to Th. Müller the observation that SOBST naturaIly supports a tensed semantics and that a tensed
semantics is needed to answer Lewis' objections.
REFERENCES
Belnap, N. (1992). Branching Space-Time. Synthese 92, 385-434.
Belnap, N. (1995). Various Notes on Outcomes in Branching Histories. 'Unpublished manuscript'.
Belnap, N. and Szab6, L. (1996). Branching Space-Time Analysis ofthe GHZ Theorem. Foundations of
Physics 26(8), 982-1002.
Eilstein, H. (2002). Against Detensers. This volurne
Heller. M. (1991). Osobliwy WszechSwiat [The SingularUniverse). Warszawa, PWN (in Polish).
Heller. M. (1988). Teoretyczne podstawy kosmologii [Theoretical Foundations of Cosmology}.
Warszawa. PWN (in Polish).
Kowalski, T. and Placek, T. (1999). Outcomes in Branching Spacetime. The British Journal for the
Philosophy of Science 51, 349-375.
Lewis, D. (1983). New Work for a Theory of Universals. Australasian Journal of Philosophy 61, 343-77.
Lewis, D. (1986). On the Plurality ofWorlds. Oxford: Basil Blackwell.
McCall, S. (1994). A Model ofthe Universe. Cambridge: Cambridge University Press.
McCall, S. (1997). Time F1ow, Non-Locality and Measurement in Quantum Mechanics. In: S.F. Savitt
(ed.) Tirne's Arrow Today. Cambridge: Cambridge University Press. pp. 155-72.
McTaggart. J.M.E (1908). The Unreality ofTime. Mind 17, 457-74.
Müller, Th. and PIacek, T. (2001). Against a MinimaIist Reading of Bell's Theorem: Lessons from Fine.
Synthese 128, 343-379.
Penrose, O. and Percival. I.C. (1962). The Direction ofTime. Proc. Phys. Soc. 79,605-16.
Placek, T. (2000a). Stochastic Outcomes in Branching Space-Time: Analysis of Bell's Theorem.
The British Journal for the Philosophy of Science, 51 (3),445-475.
Placek, T. (2ooob). Is Nature Deterministic? Krak6w: Jagiellonian University
Redhead. M. (1987). Incompleteness. Nonlocality and Realism. Oxford: Clarendon Press.
Xu. Ming (1997). Causation in Branching Time: Transitions, Events, and Causes. Synthese 112,137-92.
HELENA EILSTEIN
AGAINST DETENSERS
(Not/or tensers/
If there is Becoming,
the physicists must know it.
(Hans Reichenbach)
Abstract. The TIP dispute, between transientism and permanentism, concems the objectivity of the
concept of becoming. By detensers I mean those pennanentists who try to make their point by semantical
analysis of ·temporallanguage'. I criticize that type of argumentation as well as some other arguments
leveled by either party. If the dispute might be at least hypothetically resolved, that would have to be by
means of analyzing the basic theories of contemporary science. It turns out, however, against that
background that permanentism is irrefutable even in case it is false. The commitment to transientism is
cognitively unwarranted although it is the attitude we tend to assume in practicallife.
An old and persisting metaphysical dispute focuses on the problem whether the concept
of happening or {synonymously J becoming of events pertains to the very way of existing
of the universe; in other words, whether the way of existing of the universe consists in
the incessant succession of temporal triads whose members - that is, the respective
Past, Present and Future - objectively differ from one another.
Those who answer the above question in the affirmative are referred to in this essay
as representatives of that or another version of the metaphysical hypothesis of
transientism. Those who answer it in the negative are referred to as representatives of
pennanentism. The dispute in question is referred to here as the TIP dispute. That is
meant to be taken as a conventional name, which should not suggest that only the above
standpoints are represented in the dispute; neither should it suggest that only the above
standpoints can be consistently advocated in it. One should expect, indeed, the defense,
by some disputants, of the view that the problem cannot be solved (in the form of
a well-corroborated hypothesis) against the background of contemporary science - as
weil as of the view that the problem cannot be solved (in the form of a well-
corroborated hypothesis) within the confines of science, at any stage its development;
which for some disputants would amount to the claim that neither the affirmative nor
the negative answer can be legitimately accepted within the confines of human
knowledge. Let me state here at the outset that the view in whose favor I speak in this
93
H. Eilstein (ed.), A Collection of Polish Works on Philosophical Problems
ofTime and Spacetime, 93-126.
© 2002 Kluwer Academic Publishers.
94 HELENA EILSTEIN
paper also does not coincide with either the affinnative or the negative answer to the
above problem.
One admissible way of conceptualizing the TIP dispute is to conceive it
as concerning the acceptability of the fundamental thesis of transientism In my essay
this way of conceptualization prevails.
In the center of my interest in this essay are questions conceming the legitimacy 0/
some specijic ways 0/ conducting the dispute by some 0/ its protagonists as weIl as
conceming the real strength 0/ some arguments typically offered by either party.
In order, however, to get a necessary background for the said analysis I have first to
give abrief outline of the doctrines of transientism and permanentism.
I do it by way of paraphrasing (perhaps not entirely in accordance with the author's
intention) and developing a passage from a paper of D. Zeilicovici, who is a transientist.
The attribution of pastness operates somewhat liIce the attribution of existence in the
Kantian sense: it applies to the event camplete with a1l its properties and only defines its
status (for Kant existence. for us pastnesS).2
I believe that in order to clearly and consistently explicate his point of view it is best
for the transientist to adopt a conceptual frame where a sharp distinction is made
between the set ojproperties and the spatiotemporallocation of an individual (like an
event) on the one hand and its ontological status or, in other words, its mode 0/
existence on the other. Together with his opponents the transientist acknowledges that,
in contradistinction to individuals of the thing category (-'continuants'), the sameness
0/ an event involves the sameness 0/ the entire set 0/ its properties: events do not
undergo qualitative changes. Also, in contradistinction to individuals of the thing
category, the sameness 0/ an event involves the sameness 0/ its spatiotemporallocation
(as weIl as of any other of its relations to other events). The transientist insists, however,
that the above stipulation is compatible with the stipulation that the temporal way of
existing of the uni verse consists in events successively changing their ontological
statuses.
A difference between various versions of transientism must be taken into account
here. All transientists who in principle accept the above conceptual scheme must
acknowledge the reality - the objectivity (mind-independentness) - of futurity,
presentness and pastness as ontological statuses successively acquired by every event
(maybe with the exception of the Big Bang, if it was adefinite event). Nowadays,
however, the prevailing version of transientism is the one to which the name of
possibilism is assigned in my essay. It also comes in varieties, of which a distinction
will be drawn later on in this essay. From the point of view of possibilism futurity,
presentness and pastness should be considered derivative ontological statuses of events,
with the role of fundamental ontological statuses being assigned to open possibility,
actuality and extinguished possibility. The derivative statuses should be defined, and
their reality explained, in terms of the fundamental ones. Thus, e.g., according to some
possibilists the Future is the spatiotemporallocus of nothing but open possibilities while
the Past, with the Present as its closure, is the spatiotemporal locus of both acts and
extinguished possibilities. According to other possibilists the Future is the locus of open
possibilities, future acts and extinguished possibilities.
AOAINST DETENSERS 95
In the latter view a future act is understood as an event which is already bound to
happen at some future time - bound so, maybe (although not necessarily), due to some
respective deterministic nomic relations binding it to some events which did already
happen.
According to all classical versions of possibilism (of which more will be said
below) an event becomes present - happens - factualizes - when it is no more
simultaneous to any open possibility.
Note that in the version of possibilism which admits the existence of future acts one
has to distinguish between two kinds of change of ontological statuses of originally
open possibilities: actualization andfactualization. One, however, also has to stipulate
that for some originally open possibilities these two kinds of ontological change
temporarily coincide. Also, in a pair of such originally open possibilities which
eventually actualize and subsequently factualize, each in its due time, the temporal
order of actualization may be opposite to the temporal order of factualization.
Actualization of an event, even before its factualization, imposes the status of
extinguished possibility (in the Future) on all those formerly open possibilities whose
subsequent happening would be logically or nomologically incompatible with the
happening of the event in question. A fortiori, the happening of a hitherto open
possibility brings about the same transformation of the status of all such hitherto open
possibilities which would be incompatible in actu with the said factualized one.
To use a contemporary terminology (which is alien to the languages of common
sense and of Newtonian physics, but is useful, nevertheless, in illuminating some
aspects of these views too) - according to clossical transientism (in all its versions)
and, in particular, according to classical possibilism (in all its versions) the growing
Past and the diminishing Future constitute 'layers' of spacetime, divided by the ever
shifting Now as the closure of the Past. In this essay the term 'possibilism' refers to that
classical - 'Aristotelian' - possibilism.3 with exception of the passage where
a nonclassical version of possibilism (and thus of transientism) is briefly discussed.The
following points are of paramount importance for the proper understanding
of transientism.
First, the fundamental, from the point of view of the possibilist, ontological statuses
of events - actuality, open possibility. extinguished possibility - have to be conceived
os absolute (that is, nonrelational). The same concems the ontological statuses
acknowledged by all varieties of transientism. The ontological statuses of postness,
presentness and futurity are supposed to be had (when they are had) by particular
events in themselves, and not relative to these or other events, whatever theu
ontological statuses; not relative to instants of time (whatever one's theory of instants
may bel; not relative to anybody's knowledge or ability to acquire knowledge of the
current ontological status ofthis or that event; not relative to anybody's perception or
awareness of anything. Thus. e.g., it may be true at a given instant that a given event is,
say, present or past, but that does not mean that it is present or pastjust with respect to
that instant, while being in some other ontological status with respect to some other
instant. Indeed, while presentness is a transitory and nonrelativizable ontological status
of events, the presentness of an instant also has to be understood os its transitory and
nonrelativizable characteristics. consisting in the presentness of events with the
corresponding spatiotemporallocation.
96 HELENA EILSTEIN
The stipulation that an event becomes present when it is no more simultaneous to any
open possibility is not inconsistent at all with the stipulation about the absoluteness of
ontological statuses of events: for it still holds against the background of that stipulation
that when an event becomes present, there is nothing such that 'with respect to it' it
would be not present.
An instant may be just that instant at which an event in question is temporally
located, while some other events and other instants preced or succeed that event in time
- thus belonging, respective1y, to the Past or to the Future when the event in question
is present - but the transitory, instantaneous ontological status of presentness of the
event in question cannot be defined in terms of any temporal relations, like simultaneity
and precedence, and that accords with the nonrelational character of that ontological
status.
Second, in the view of transientism no event can at any time be in none of the above
ontological states. The transitions between them have a character of unanalyzable
'timeless' jumps.
Any doctrine which does not incorporate the above two transientistic stipulations is
not a bona fide transientism, even if its representatives, as it often indeed happens
among the protagonists of the TIP dispute, pay a lip service to transientism.
Let me make explicit a point inherent in the above considerations.
The concept of the growing Past and diminishing Future as two 'layers' partitioning
spacetime across the temporal dimension, with the shifting Present as the closure of the
Past, is characteristic for classical transientism. It was easy for that doctrine to see itself
in concord with Newtonian physics (though the latter by no means implies it). Indeed,
against the background of the Newtonian views on space and time - or, to use again
contemporary terminology, against the background of the Newtonian theory of
spacetime, conceived as the uniquely foliated continuum of successive instants
(instantaneous three-dimensional spaces) - it was natural for the transientist to
stipulate that the successive positions of the shifting Now partition the totality of events
of all ontological statuses into classes of genuine - absolute - simultaneity, which
were supposed to correspond to the Newtonian instants of time. The partition of the
totality of events according to the relation of Copresentness had to coincide with the
partition of that totality into the classes of genuine simultaneity; where Copresentness is
to be understood (in this conceptual scheme) as the time-invariant relation between
events which belong or belonged or will belong to the same Now. 4 From the
contemporary point of view the above stipulation associated transientism with the idea
of a privileged stratification of spacetime into instants. The questions whether the
contemporary transientist has a reason to look for a way to disentangle his doctrine
from the association with that idea; and in case he does, can he do it in an
unobjectionable way, are arduously disputed at the contemporary stage of the TIP
dispute. The problem is particularly acute for possibilistic transientism.
Let me proceed with my characteristic of transientism. The supporter of this
doctrine who accepts the conceptual frame of ontological modes of existence should
also, I believe, admit there the concept of nontransitory Existence of events or, in other
words, of nontransitory Occurrence of events - or, in still other words, the concept of
nontransitory Reality of events. That would result in the admission not only of tensed,
but also of tenseless forms (modes) of verbs, including the verb 'to Exist' into his
AOAINST DETENSERS 97
specific language. (As it seems natural for the user of an Indo-European language, in
this text the tenseless form of a verb will be homonymous with the form of the present
tense). 'To Exist', with reference to events - or 'to Oeeur', as I will prefer to speak
- would mean for the transientist to be in any of the above, in general transitory,
modes of existence.
Transientism, of course, is obliged to ans wer the fundamental question: what does
the transience of events consist in and on the basis of what evidence should we assume
that the concept of the transient Now has an objective significance.
Transientists give a number of answers to that question. Different answers do not
necessarily contradict one another; they also may complement one another.
According to some transientists (e.g. C.D. Broad) the concept of becorning just
corresponds to the most fundamental feature of the temporal aspect of the way of
existing of the universe. Accordingly, it is essentially indefinable. It owes its
meaningfulness to its being directly rooted in our experience, particularly in our basic,
everyday, prescientific experience -let us call it the primary experienee.
It seems to me - a1though I am unable to support my view with a reference to
corresponding explicit assertions - that in the view of some thinkers the transience of
events is a 'result' of the effieaey 0/ causes. According to this view causes 'produee'
their effeets - or, synonymously, 'precipitate them', 'make them to happen at a due
time'. Even if at a given time a given event enjoys the status of a future act, it must yet
be 'made' to 'Jaetualize' (that is, to happen); and it is the 'way of operation of causes'
which has to take the responsibility for the 'Iapse of time' between an event 'becorning
inevitable' and that event 'happening'. It is sometimes c1aimed that the efficacy of
causes is directly manifested in our primary experiences, like the experience of being
pushed or pulled. The need to resort to dynamieal eoneepts in our description of
physical processes and nornic relations between them also may be supposed by some to
be a decisive evidence in favor of the efficacy concept of causation.
The efficacy theory of causation seems to me responsible for the fact that
transientism is not neeessarily associated with possibilism. Indeed, in the epoch of
c1assical physics scientists typically took both transientism and deterrninism for granted.
The Future represented for them the collection of all those events which were ever
bound to happen at the due time, but were not yet 'made' to factualize.
In the possibilistic version of the efficacy theory of causation causes are thought of
as endowed with various degrees of propensity to produce these or other of their
alternative possible effects. Note however that the possibilist does not need to comrnit
himself to the efficacy theory of causation. He may maintain that regardless of whether
the efficacy theory of causation is adequate, the flow of time 'basically consist' in the
successive actualizations and corresponding attendant extinctions of originally open
possibilities.
Now however, at the end of this long march to the main topic of my essay, let me
shortly present the basic thesis of permanentism.
Cast in the specific language of the transientist (which is richer, as far as the
problems pertaining to the nature of time are concemed, than the specific language of
the permanentist; and which, accordingly, provides one with means to explicitly
formulate both contending metaphysical hypotheses), permanentism amounts to the
claim that the universe is nothing but the totality 0/ Existing aets which may - and in
98 HELENA EILSTEIN
In the conceptual frame of the permanentist 'laws of nature are simply codifications
of certain deep regularities,.8 There of course is no application there for the idea of the
'production' of effects by their causes. The permanentist, and the opponent of the
efficacy theory in general, has to grope for an adequate version of the opposite,
regularity theory of causation. He interprets the dynamical variables as parameters
in the corresponding Iaws of physics, and thus assigns them an important role
in characteristic of events, without, however, any reference to the concept of efficacy.
In the terminology of this essay the name of detensers and the correlated name of
tensers is used in a narrower sense than the above custom dictates. The first one is
employed with reference to representatives of abrand of permanentism. Detensers, in
my tongue, are those among them who propose to consider the TIP controversy in
a specific problem perspective; that is, who represent a specific attitude conceming the
question that has been mentioned at the end of the preceding paragraph. The name of
detensers and, respectively, of tensers, is also employed here with reference
to participants in the TIP dispute insofar as they in their particular contributions do not
transcend the problem perspective of detensers in the above hinted sense of the word.
Some of these authors transcend that problem perspective in their other contributions or
parts of contributions. But some transientists are genuine tensers, as they indeed share
the detensers' problem perspective. They do it because they share the views
of detensers on, on the one hand, the signijicance for the dispute of such utterances,
ways of thinking, feelings and practice of people (particularly in our culture) that are
characteristic for the scientijically unsophisticated common sense; and, on the other
hand, the signijicance that should be attributed in the dispute to prevailing fundamental
scientijic theories, their contending interpretations, as weil as to conjectures discussed
in contemporary science.
In this part of my essay and in the next one my attention is directed towards the
detensers. I refer to the polemics between them and the tensers as to the D/f dispute. \0
The approach of detensers is thusly presented by one of its representatives, L.N.
Oaklander:
Although the issue is metaphysical, the dispute between tensed and the tenseless views has,
until quite recently, centered around temporal language.' ,
The intention of Oaklander and other detensers (as weIl as of tensers) is to continue
that tradition. The general method employed by detensers in order to secure the victory
for permanentism consists in analyzing semantical relations, which according to them
obtain between sentences pertaining to time. It consists, namely, in attempts to show
that it is possible to perform this or another kind of 'reduction' of all tokens
of sentences containing expressions associated with the specific conceptual frame
of transientism (that is, all tokens of the so-called tensed sentences) to tenseless
sentences, where these expressions would be substituted by ones from the language
of permanentism. The reductions are supposed to assign to every token of a tensed
sentence a tenseless sentence that would convey the total information about the
objective reality conveyed by means of the given tensed sentence-token. 12 That should
elicit the sufficient conditions for the truth of respective tensed sentence tokens, but at
the same time also prove that the objectivity of the transientistic ontological statuses
of events by no means is necessary for the truth of the respective tensed sentence
tokens. Consequently, that also should show the eliminability of the specific
expressions of the language of transientism from the language of the purely cognitive
discourse 13
Detensers use to stress the alleged analogy between their semantical operations and
the operation of substituting, in the cognitive discourse, the reference to spatial
coordinates of events in some chosen coordinate system for the subjective indexicals
like 'he re , or 'to the left of me '. 14
100 HELENA EILSTEIN
The attitude of detensers in the T/P dispute thus radically differs from the attitude of
those permanentist who admit that transientists offer us a specijic son of information
about the objective reaIity - an information which cannot be cast in terms of a
permanentistic language - but who at the same time either insist that this information
is demonstrably false or claim that it is implausible in the light of our empirical
knowledge.
Oaklander's belief conceming the extant shape of the T/P dispute is erroneous.
In accordance to what I stated above, it is easy to name a number of participants in that
dispute who have not conducted it in the guise of the DIT dispute. The names of
A. Einstein, K. Gödel, H. Reichenbach, A. Grünbaum, M. Capek, S. McCall, H. Stapp
and many other can be used for exemplification of this point.
Q. Smith is closer to truth than Oaklander when he states:
[T]he issue between the defenders ofthe B-theory [viz. pennanentists] and the defenders of
the A-theory [viz. transientists] is of fundamental ontologicaJ importance. But analytic
philosophers discuss this issue almost excIusively in tenns of the Ianguage we use to
describe the temporal detenninations of events. Theyengage in what Quine calls 'semantic
ascent' , that is, they redirect their concem from the things themselves to the words we use
to describe things.'s
In the view of this assertion the attempt to conceptualize the T/P dispute as the DfI'
dispute is a relic of the approach of analytic philosophy to ontological problems.
As to the enterprise of detensers, it is obvious that since meanings of expressions
depend on the respective language, semantical relations also rnay hold within some
language and at the same time fail to hold between the homonyms of the sentences in
question in another language. The semantical reductions of detensers actually are made
with reference to the meanings of the respective sentences and sentence-tokens in that
or another version of the specijic language of permanentism. Therefore whatever their
variety, the undenaken reductions invariantly prove to be successful. However, the
achievements of that labor cannot be more relevant to the T/P dispute than the
'victories' achieved by a boxer over his exercise bag are for his championship in
a competition - unless the detensers convince their opponents or the witnesses of the
dispute that it is indeed an ontologically justified thing to restriet oneself to the specific
language of permanentism in one's attempts to account for the way of existing of the
objective reality.
The detensers indeed believe that their reductions are ones with respect to just that
language which according to their views actually matters in the dispute - namely our
everyday life language: the one which corresponds to 'common sense'. According
to N. Oaklander, it does not follow from the detenser's theory of time and tense that our
[ordinary, common sense] intuitions about time are mistaken (although they need to be
properly interpreted) or that tensed language is false. Indeed, once our ordinary
language of time is distinguished from a metaphysically persr.icuous language of time,
it can be shown how and why our temporal intuitions are true. 6
In the view of this assertion the task of the detenser does not consist in a critical
exarnination of the common sense intuitions about time. It rather consists in showing
how, in cases there is a need for that, these intuitions can be expressed in
a 'metaphysically perspicuous' way.
AGAINST DETENSERS 101
What is, in the views of the protagonists of the Dff debate, the alleged specifically
philosophical way of cognition of the fundamental features of the objective reality? For
most of them it evidently consist in the analysis of the wisdom of the tenets of common
sense. Accordingly, while the reflections on Relativity, relativistic cosmology,
Quantum Mechanics and the applicability of the prevailing concepts of spacetime at the
Planck' s scale play an essential role in the TIP dispute in general, the protagonists of the
Dff debate detach themselves from these matters.
There is another, rather strange, feature of the attitudes of the protagonists in the
Dff debate. Their attention is exclusively directed toward the problems of
objectivity of those ontological statuses of events which constituted the matter of
contention between J. McTaggart and his early polemists (like the permanentist
Russell, the transientist Broad). The approach of these thinkers to the TIP dispute
was formed in the epoch when fundamental theories of physics were consistent with
determinism, and philosophers - at least those adverse to irrationalism - were in
the grip of deterministic prejudice. 20 Accordingly to that peculiar limitation of the
problem horizon of detensers, their efforts to 'reduce' transientistic statements turn
out to be directed towards those which contain references to what from the point of
view of possibilism are merely 'derivative' ontological statuses of events. Detensers
seem to be unaware of the role of possibilism in the TIP dispute and tensers do not
object to that peculiar approach to the problem of 'reduction'.
102 HELENA EILSTEIN
3. DETENSERS AT WORK.
An analysis of some examples of the detensers' 'reductions' should illustrate the failure
of their enterprise. Nowadays detensers advocate 'the New Theory of Time'.
It differs from their 'Old Theory' in that its supporters reJect the program of assigning
to every token of a tensed sentence its tenseless synonym. I They came to the view that
in the language which is identified by them as the language of common sense the
relation of synonymity never holds between a taken of a tensed sentence and a tenseless
sentence. They propose instead two basic methods of reduction, in the sense of
assigning to tokens of tensed sentences their tenseless counterparts which are supposed
to make explicit the truth-conditions for the respective tensed sentences, and thus also
disclose their objective content.
D.H. Mellor represents one of those methods:
[L]et... S be any token of 'It is now 1980' ... S is true if and only if it Occurs in 1980.22
All that may be so in the language of Mellor, but surely it is not so in the language
of the transientist; not so in the language of common sense. In the latter language S is
true if and only if the events which are happening (or, to indulge a redundancy, which
are happening just now) have a date within the year 1980. According to the transientist,
Mellor's assertion only shows that he fails to understand what 'now' means. Indeed, the
truth-conditions af a sentence or sentence-token that is possessed of adefinite truth-
value are determined by its meaning in a given context, and the meaning obviously is
language-dependent. Mellor attributes to his S a meaning it only may possess in
a language that corresponds to the conceptual frame where the concept of happening,
becoming, passing into the Past is not represented at all.
Whether or not there is some stare of affairs in the world can hardly be settled by whether or
not we have linguistic devices to pick it out24
If there is no real happening, then no taken of the sentence 'It is now 1980' has
a legitimate place in a positive descriptian ofthe objective reality, because none ofthese
tokens are true.
In order to make my point more salient let us consider a token N of the sentence 't is
now', where t is a time coordinate. According to Mellor, N Is true if and only
if it Occurs at t. But according to the rules of the language of the transientist N is true if
and only if all events which are 'just' happening have the coordinate t. As a necessary
condition for N to be true, there must be a coordinate system whose classes
of simultaneity of events are classes of their Copresentness. 25
Another method of reduction of tensed sentences proposed by a number
of detensers employs the concept of coreporting.
[T)wo sentences can express different propositions and yet report one and the same event
or state of affairs; e.g. "Ibis is water' and "Ibis is a collection of H 20 molecules'... report
AOAINST DETENSERS 103
the same state of affairs ... It could he claimed that the same holds for the appropriate use of
indexical [tensed] and nonindexical [tenseless] sentences; the tokening at '70f 'Oeorgie
flies at this time (at present)' is coreporting with the non-synonymous 'Georgie Flies
[capitaI letter mine - H.E.] (tenselessly) at '7.' since Oeorgie's flying at this time is the
same event as Georgie's Flying at '7 given that this time is l? This effects the ... ontological
reduction of the becoming of events to their hearing temporal relations to each other.26
That is a glib talk. Georgie's flying reported at t7 in the sentence token f, 'Georgie
flies presently' may be 'the same event' as Georgie's Flying at t7 reported in the
sentence F, 'Georgie Flies at t/. This notwithstanding, from the point of view of the
transientist these two sentences do not convey the same information about the objective
reality. The sentencefsays that an instantaneous stage of Georgie's flying is 'just now'
undergoing the transition to the ontological status of pastness. The sentence F, on the
contrary, does not bring the information about the current nowness of the respective
stage of Georgie's flying.
The transientist may offer an analogy here (to be sure, it would be a fitting one only
from his point of view): The inhabitants of the Valley of the Blind, who do not believe
tales of a stranger about there being in the objective reality something he calls 'colours',
can convey a lot of information about a given rose flower by means of their language;
but although they speak of the same flower to which the stranger refers, no sentence
produced by them would convey the information about the redness of that flower.
A more radical interpretation of the 'reduction' offto F would present the transition
from the first of these sentences to the other as a kind of 'rational reconstruction' which
preserves the real information be it adequate or inadequate
- inherent in f while doing away with meaningless expressions. Some remarks by
Earman and Gale suggest such an interpretation. Indeed the authors question the
meaningfulness of the specific expressions from the language of transientism. They find
these expressions 'mysterious', particularly as long as it is maintained that some of
them defy semantical analysis, so that their meaning has to be grasped intuitively, on
the basis of the common human experience. For a reason which is mysterious to me the
authors seem to believe that in order to endow his specific expressions with a meaning
the transientist would have to commit himself to the belief in a 'mysterious Mr. X out
there' doing 'The Shift' .27 It seems to me obvious, however, that people who use
everyday life language associate some definite meanings with the specifically
transientistic expressions regardless of whether these or other of them believe in
'Mr. X'. They do it even if it be the case that they cannot speIl the meanings of these
expressions by explicit definitions - and they do it despite the fact that the expressions
in question cannot be defined by means of the language of permanentism.
I see the upshot of my considerations in this section in the conclusion that all
manners of 'reduction' undertaken by the detensers can be convincing only for them.
If the protagonists of the Dff dispute wish to constructively participate in the T/P
dispute, they should perform a 'semantical descent' and quit trying 'to redirect the
concem' of the protagonists of the TIP dispute from 'things thernselves' to 'the words
we use to describe things' .
104 HELENA EILSTEIN
All my above criticism does not mean that the detensers fail to bring to the
philosophical forum anything deserving consideration. On the contrary, I believe they
help, albeit inadvertently, to draw our attention to an important issue, one, however, that
belongs in the domain of philosophical anthropology rather than philosophy of time.
The issue concerns the relation between the contents of consciousness of Man in the
purely cognitive mode of his mind and the contents of consciousness of Man as the
emotional-volitional subject. In other words, the issue concerns the relation between the
ideas Man is able to conceive and critically examine when his mind operates on the
level of sophisticated cognition (brought by scientific, logical and philosophical
intellectual activity) - and ideas (be they true or false) in whose grip his rnind
is caught when his nature of emotional-volitional subject comes to the fore and
influences his thinking.
Detensers inadvertently draw our attention to this topic. As it has been presented
above, they maintain that despite the falsity of transientism some tokens of tensed
sentences must be true. They see the justification for this claim in that albeit tokens of
tensed sentences are (as they believe) eliminable from the purely cognitive discourse,
they prove to be 'indispensable' in the human discourse (and verbal thinking) in
general. There is, the detensers point out, an ineradicable need in human life for 'tensed
beliefs,.28 Specifically, they point out, tensed beliefs and sentence-tokens which express
them are indispensable for the human timely action. Thus, at the time of the
spatiotemporal Location of the Beginning of Boiling of milk in a kettle on the range in
Mary's kitchen, Mary Must believe that the milk is starting to boil now, unless she Fails
to turn off the flame and Save her breakfast. lt is irrelevant for that feat whether
simultaneously (in her system of coordinates) with the milk Starting to Boil she Knows
what Is the clock time of the milk Starting To Boi!. It is only important for her to Have
simultaneously with that phenomenon an 'it-is-happening-now' belief concerning it.
The linchpin of the detensers' attitude to the question of truth-values of tokens of tensed
sentences consist in the assumption that the indispensable role played by the appropriate
beliefs and corresponding sentence tokens with respect to the human 'timely actions'
warrants that they cannot be all untrue. But as we have seen above, as long as the
detenser does not tinker with the ordinary way of understanding the tensed sentence-
tokens, he has to commit himself to the claim that no tokens of tensed sentences are
true. He thus should find hirnself perplexed.
That, however, is so because the detenser does not appreciate the profundity of the
difference between the world-view of the perrnanentist and the one of common sense.
He tries to preserve in his conceptual frame the common sense concept of the agent.
The latter, however, presupposes the concept of the efficient cause, which can only
belong to the conceptual frame of transientism. In the universe in which all events
simply Are, no events are brought into existence or factuality by anybody's 'timely
action' - although some processes (causal chains of the respective events) which are
conceived by common sense as 'timely actions' ofrespective individuals may belong to
nomic necessary conditions for the Occurrence of some events. From the point of view
of the permanentist 'free and efficient agency' is a 'stubborn illusion' which
characterizes the intensions of mental states of those people who attribute the role of an
AGAINST DETENSERS lOS
primo, that (in accordance with the views of a number of authors) libertarianism is a
self-contradictory doctrine, which, however (in accordance with the view of Peter
Strawson) is virtually irresistible for our mind at times when we indulge in the attitudes
of common sense instead of philosophizing. There is no place here, however, to
elaborate on this topic.
present - will become some other moment of time and thus cease to be identical with
itself!
Here, too, the blunder of the presented reasoning is obvious. In the language of the
transientist the term The Present' (or 'Now') refers to the collection of events which are
just happening. Since an event rnay be in this ontological status only transitorily, the
name in question is successively inherited by ever new collections of events. There is
an obvious analogy here to the name 'municipal council' which refers to an ever new
team of people in a city whose statute prohibits a person to be elected to the council
more than once.
A popular argument among permanentists, also reproduced in the paper of Earman
and Gale33 consists in charging their opponents with being unable to escape the
'absurdity' of the claim that 'time passes or shifts at the rate of one second per second'.
The prernise of that argument consists in the stipulation that
[i)f events are to change from being future to present and from present to past, they must da
so in relation [italics mine, H.E.] to some mysterious transcendent entity, since temporal
relations between events andlor times do not change.
This premise, which also is the linchpin of the McTaggart's argument against
transientism (or, in the words of McTaggart, outright against the 'reality of time')
manifests a total failure to understand the gis~ of transientism. As it has been elaborated
on earlier in this essay, according to transientists events change their ontological
statuses absolutely, and not with respect to anything. 34 They as a rule use the expression
'flow 0/ time' in a metaphorical sense, meaning the successive happenings of events.
A river flows at a definite rate with respect to its banks, but it does not rnake sense to
stretch the metaphor too far and to ask what is that 'transcendent', 'fixed' entity with
respect to which time itself is supposed to flow at this or that rate. By the way, the
metaphorical expression 'flow 0/ time' rnay be attributed a meaning with respect to
which the statement 'time flows at the rate %ne second per second' turns out to be
a triviality rather than an absurdity. For, the standard transientist would say, if two
events are Divided in time by the distance of one second, then, if one of them happens
at a time t, then, if the other also happens, it does so one second later.
I assume that both for transientism and for permanentism the internal inconsistency
neither has been nor can be proved. Accordingly, if it be possible to find legitimate
arguments for or against either of theID, then it must be on the basis of experience.
The turn to everyday life experience. As it has been stated above, according to some
transientists their thesis receives a decisive support from the analysis of the character of
the everyday, 'primary', experience. It seems unquestionable, indeed, that as far as the
way of our perceiving the uni verse is concerned, transientists have an impressive point
in their favor. That is explicitly avowed by some permanentists. H. Weyl, whose
formulation of the thesis of permanentism was quoted above, continued his assertion
the following way:
Only 10 the gaze of my consciousness, crawling upward aIong the life-line of my body,
does a section of this world come to life as a fleeting image in space which continuously
changes in time.
It is seen from Weyl's metaphorical formulation that in his conceptual scheme the
consciousness of an episternic subject is represented by amental process supervening
108 HELENA EILSTEIN
upon a physical process and therefore associated with adefinite worldtube. Without
busying ourselves with the intricacies of the Mind/Body problem we may satisfy
ourselves that the consciousness of a human episternic subject is thus represented in
Weyl's conceptual scheme by a causal chain - though by no means an isolated causal
chain. An important property of that chain consists in that its successive links (states of
consciousness) are characterized by intensions which feature, in a specific way, both
retention and change of components of intensions of the preceding stages. To that
consciousness, Weyl says, the perceptually accessible section of the world 'comes to
life' as a fleeting image in time. That also is acknowledged in the above quoted
statement of K. Gödel. A point of contention, however, consists in whether from the
assumption that the universe appears to be that way to the perceiving subject one has to
conclude that the transientist is right in his views conceming the objective reality?5
Upon a consideration, however, one finds, I believe, that the 'total fit' between
transientism and the character of our perceptions also is questionable. One seerns to
observe the temporal flow of reality but not happening of events, not events while they
are present. It is weIl known that the attribution of presentness by a scientifically
untutored subject to some 'events' (- actuaIly, to processes whose temporal extension
is overlooked in our perception) is often based on illusion, while the 'events' in
question precede their perception by long intervals of time. The transientist at most
rnight claim that we perceive adequately enough the happening of perceptible 'extemal'
events in our close spatiotemporal vicinity and also are aware of the presentness of our
own psychical events (inasmuch as they are accessible to our awareness). That,
however, also would not be entire1y adequate. For, strictly speaking, lirnited speed of
physical signals both outside our brain-cum-sensory-apparatus and within it results in
that whatever we perceive in the outside world must be, from the transientistic point of
view, past; and our perceptions also must be past when we become aware of them.
When we focus our attention on a particular 'event' and perceive it as 'happening
before our very eles' we only manifest, firstly, our inability to perceptually distinguish
a genuine evene from a process endowed with a very short temporal extension; and
secondly, our inability to distinguish in our perception a 'very near Past' from the
'Present'. We also manifest, moreover, our inability to distinguish the very fresh
memory from what rnight be called an instantaneous perception. As a result the section
of the uni verse which is perceptually accessible to us 'comes to life to the gaze of our
consciousness' as a continuous temporal flow, and not as a succession of distinctive
happenings of events.
Bergsonians, like Capek, make a lot of this feature of continuity in the way we
perceive the uni verse. In their view, it discloses to us the genuine character of time.
And yet, if one wants to have the idea of distinct ontological statuses of events in his
conceptual frame, then one has to conceive the happenings of events as unalyzable
'jumps'. One has to take into account 'the atomicity ofbecoming'.
The question arises conceming the compatibility of transientism with the concept of
continuity of spacetime and of processes.
The element of 'atomicity' appears in the nowadays standard Cantorean conception
of a continuum as a set of individuals (points) whose cardinal number is c.
The Cantorean continuum is intrinsically 'pulverized' into points.
AGAINST DETENSERS 109
But surely the spacetime points would then only 'emerge' as idealizations of minuscule
areas with blurred boundaries and not as genuinely distinct Cantorean points.
Bergsonians might consider this change in the approach to the nature of continuum as
water for their mill. And yet, transientists are the ones for whom that concept of
continuum, if applied to time and processes, speIls trouble, while permanentists are able
to accommodate it into their model of the universe. Bergsonians run into the snare
transientists always strived to avoid. 40 Namely, against the background ofthat approach
to the nature of continuum, in order to make sense of their doctrine, they seem to need
to introduce a kind of transcendent static substratum or object with respect to which
their time flows. Indeed, we have here the 'flow of time'; where, however, is the
Present? And with its disappearance, how could the Past be different from the Future?
There seems to be a need for an extraneous 'marker' which would have to be passed by
an event in order to slip into the Past. And how can this idea of the time continuum be
reconciled with the idea of open possibilities in the Future - the idea which is so dear to
Bergsonians? A 'continuous', not all-at-once transition from possibility to actuality
seems to be inconceivable (even if the probability of actualization of an event may
change continuously), and thus the shifting division between the Future, the locus of
open possibilities, and the Past has to be Sharp.41
It seems to appear from my above considerations that only these versions of the
theory of becoming which dissociate the transient time as perceived by us from the
transient time as it is are consistent. The version which fits best the nature of OUT
perception does not represent a consistent view on the objective nature of time.
It is necessary fOT the protagonists of the T/P dispute to transcend the horizon of
'direct' perception and to locus their attention on basic theories 01 contemporary
science as well as on conjectures nowadays discussed in it.
symptoms of everything that is happening at t in the uni verse - though these symptoms
may be not cognitively available for such imperfect subjects of empirical cognition as
we are).
For the sake of our analogy, let us, further, assurne that transspatial determinism is
true. Consider the state of the uni verse at the present instant of time. Let it have the
coordinate t in our frame of reference. It would be then the claim of Aristotelian
possibilism that there are no more open possibilities at t. The state of the uni verse at t is
entirely 'fixed'.
To nourish our imagination, assurne that the closed area of our interest is a prison cell in
which an inmate is confined. At the present instant he is thinking about his wife,
craving to know whether she is alive. According to transspatial determinism, from the
purely ontological point of view he rnight be able to achieve that information by a
careful exarnination of the current state of the inside of his cell. That is, he would be
able to perform this feat if he only were a good enough epistemic subject: if he knew the
respective synchronie deterministic laws of nature and was able to gather sufficiently
complete and precise data from his closed environment and to process them.
However, according to our best knowledge, transspatial determinism is nothing but
fantasy. The poor prisoner cannot satisfy his desire not because he is not a good enough
episternic subject, but because there do not exist those synchronie deterministic laws of
nature whose existence is presupposed by transspatial determinism.
We may take it for granted that the contemporary possibilist rejects transspatial
determinism. According to the point of view of the Aristotelian possibilist, despite the
fact that all events whose date of happening is t (that is, which are happening just now)
already are actual, none of them has such sufficient conditions of their actuality as are
stipulated by transspatial determinism. The other way round, according to Aristotelian
possibilism the fact that events happening at t do not have such sufficient conditions of
their actuality as are stipulated by transspatial determinism, does not prevent all events
happening at t to be actual.
Our analogy is supposed to illustrate the truth of the thesis that the concept of
actuality does not need to be based on the idea of nornic sufficient conditioning of this
or that sort. And that is the reason why indeterminism is compatible with actualism and
thus also with permanentism.
Before finishing this section it is worthwhile to devote some attention to the concept
of probability. Although probability was introduced to physics by deterministic
thinkers, indeterminism does not deny that at least some events in whose case the
deterministic supposition fails may be subrnitted to laws according to which
probabilities of their happening (if transientism is true) or simply of their Occurrence at
a given spatiotemporal location (if permanentism is true) are determined by some
respective actual situations that Precede them.
The interpretation of the concept of probability is one of the keenly disputed matters
in contemporary science-oriented philosophy. While some views on the nature of
probability (namely, the propensity interpretations) presuppose possibilism, there are
also such ones (namely, versions of the jrequency interpretation of probability) which
are compatible with actualism and, in particular, with permanentism. They are available
for possibilists, too, and square weil with the fact that the concept of probability is used
not only in predictions. The possibilist does not have to deny that in some contexts one
AGAINST DETENSERS 113
has to apply a concept of probability that is not associated with the idea of propensity or
with the idea of efficient cause in general. Thus, e.g., let us consider (from his point of
view) the case of the poor prisoner who is presently thinking of his wife. The wife is
either alive or dead and he cannot know which is the case. He may, however, be in
pos session of some data and some nomological knowledge that enable hirn to
determine whether the probability of her staying alive is or is not greater than the
probability of her being dead.
Let us assign the name of probabilism to the thesis that there are objective
probabilistic laws. In the view of their neutrality in the TIP dispute, the support
indeterminism anti probabilism nowadays appear to obtain /rom empirical science
cannot automatically count as a support for transientism.
References to Relativity playaparamount role in the current stage of the TIP dispute.
Remarkably, too many protagonists think it is sufficient to take into account just that
theory of time which is germane to the standard interpretation of Special Relativity.
That, however, outlines a too narrow problem horizon.
As a matter of fact, the very concept of time tums out to be questionable in
contemporary physics.
According to some physicists who work at the frontiers of current science it loses its
validity as physics descends to the more and more deep levels of rnicroworld. Thus,
e.g., in his paper on distinct meanings of the notion of time C. Rovelli points out, that
[i]n certain fonnulations of quantum gmvity a concept of time is entirely absent, and hope
is expressed that time could be recovered within approximations"
Rovelli. in the series of papers ... has not only propagated the idea that at the fundamental
level a well-defined concept of time is totally absent. but also has obtained some
interesting results showing how could physics be done without the usual notion of time45
The same idea also is presented in the paper of M. Heller in this collection.
This would undermine the concept of the objectivity of the 'transient Now'. It does not
make sense to speak about there being no objective happening at the lowest 'layers' of
the structure of the uni verse, and of the 'flow of time' somehow emerging at a higher
level of that structure. According to S. Hawking
[i]n the very early universe. when space was [the pennanentist should mther use the
expression 'Is' - H.E.] very compressed. the smearing effect of the uncertainty principle
can change ... [the] ... basic distinction between space and time... [1bat would be so] under
some circumstances. When this is the case ... we might say that time becomes fully
spatiaIized - and it is then more accumte to taIk not of spacetime but of a four-
dimensional space. Calculations suggest !hat this state of affairs cannot be avoided when
one considers the geometry of the uni verse during the first minute fraction of a second.
The question then arises as to the geometry of the four-dimensional space which has to
somehow smoothly join onto the more familiar spacetime once the quantum smearing
effects subside.46
universe - the closed Friedman world (corresponding to the popular idea of a world
with a 'Big Bang' and a 'Big Crunch'):
The following intetpretative eomment [whieh is only eonsistent with the mathematical
strueture of the eonsidered model, but not implied by it; see the Footnote 10 to the quoted
paper] illustrates the situation: For beings Iiving inside the c10sed Friedman model [it is the
ease that by] studying eosmology they ean leam about the existenee of the initial
singularity in their past and they ean prediet the final singularity in their future. Neither of
these singularities is direetly aeeessible to them; however, they have leamed about the
singularities by eolleeting information from within spaeetime in whieh they live ... Suppose
further that the world under eonsideration has been ereated by a Demiurge in the initial
singularity ... (It turns out for mathematieal reasons that] for the Demiurge the beginning of
the world is simultaneously its end. (That is, from his 'outside' point of view the total
history ofthe world eollapses into a spacetirne point].
To which the author adds (directly below the quoted passage) the additional remark:
'Theologians always claimed that God is atemporal and therefore everything happens
instantaneously for God'. Leaving aside the last remark, which corresponds to the
metaphysical commitments of the author, we can conclude that one can attribute two
'aspects' to the closed Friedman world. Regarded from 'inside' it is perceived as
extended in time and representing adefinite succession of stages. Regarded
'holistically', as if 'from outside', however, it seems to reduce to a 'spacetime point' in
which aB these stages are present 'at once' and with respect to which, thus, the concept
of 'time flow' appears utterly inapplicable.
To proceed with examples of challenge to the transientistic approach to time by
a number of contemporary physicists - in his book The Fahrie o[ Reality David
Deutsch denies the universal applicability of the concept o[ the arrow o[ time (also in its
relativistic interpretation) to the description of the totality of physical phenomena:
[T]he approximation that time is a sequenee of moments ... must break down badly in
eertain types of physical processes .... Aeeording to quantum physies, as best as we ean
tell ... the earliest moments which are, to a good approximation, sequential, oceur roughly
when c1assical physics would extrapolate that the Big bang had happened 1043 seconds ....
earlier... A second and similar sort of the breakdown of the sequenee of time is thought to
oceur in the interiors of blaek holes and at the final eollapse of the universe .. .if there is one.
In both eases ... the ... gravitational forees tear the fabrie of spaeetime apart ... Thirdly, it is
thought that on a sub-mieroseopie seale quantum effeets again wRtp and tear the fabrie of
spaeetime, and that c10sed loops of time ... exist on that seale ... [Possibly t]his sort of
break~~wn of the sequenee of time [also] oceurs near sueh objects as rotating blaek
holes.
All these findings and conjectures are consistent with permanentism. They are not,
however, consistent with transientism in whose views there must be a radical distinction
between space and time and, as far the classical transientism is concerned, there also
must be an objective and ubiquitous arrow of time pointing out from the Past to the
Future.
It is worthwhile to mention here the contemporary disputes conceming the
possibility of time trave!. Notably General Relativity presents one with a class of
models of the uni verse and thus enjoins scientists to consider which of them
corresponds best to the observed reality. Some of these models admit CTCs - that is,
closed timelike eurves (over cosmological distances). Some authors do not rule out that
the real universe satisfies such a model; some of them, moreover, believe in the
AGAINST DETENSERS 115
possibility of time travei 0/ human individuals or at least do not rule out such a
possibility. It has been explained in a number of works (e.g. by J. Earman) that the idea
of time travel not necessarily is pregnant with inconsistencies, like the notorious
'grandfather paradox' .49
Note one consequence of the conjecture about the possibility of travelling into one's
past. While the time traveler feels that he came "into the Past", which is to some extent
known to hirn from records, those receiving his visit feel that he arrived "now". In this
way the concept of Now becomes relativized to particular persons, losing the character
attributed to it by transientism.
Pace Earman, time travel should be considered a specific kind of backward
causation. lassume that the representatives of classical transientism oppose all versions
of the idea of backward causation, particularly if they accept the efficacy conception of
causation. Even if there are future acts, neither they nor the open possibilities are
supposed to play the role of succeeding causes of present or past acts. Though the idea
of backward causation is also opposed by some permanentists, there is nothing in
permanentism as such that would prevent its acceptance. It seems worthwhile to
mention some other conjectures in contemporary physics conceming backward
causation. The talk about backward causation e.g. arises in the context of the conceptual
puzzles of Quantum Theory. For O. Costa de Beauregard his reflections on the EPR
problem have played an essential role in renouncing the 'assumption that advanced
actions are excluded'.5O Some other authors, like P. Dowe or H. Price also see in
backward causation the solution of the EPR riddle.
Let us, however, leave the above mentioned problems aside and focus on these
relativistic models of the universe which conceive it as admitting (at all levels of its
structure) stratifications of the spacetime into spacelike hypersurfaces with no causal
loops and no obvious contradiction with the assumption of definability, in empirical
terms, of the ubiquitous time's arrow (that is, the relation of objective and absolute
temporal succession, prevailing at least in pairs of absolutely nonsimultaneous events).
Against this basis, let us take up the much discussed question of the significance of
Relativity for the TIP dispute.
As I have mentioned above, it is Special Relativity in its standard interpretation that
attracts much attention of the protagonists.
According to that standard interpretation it is an essential feature of Special
Relativity that it denies the existence 0/ a uniquely privileged stratification 0/ the
spacetime into a set 0/ spatially ubiquitous spacelike hypersurfaces. It maintains that
neither the nomic order of the uni verse, nor some global conditions prevailing in it
bestow such a unique status on any of the conceptually possible stratifications.
Accordingly, in this interpretation Special Relativity rejects the concept of absolute
simultaneity. And in this way it totally undermines the idea of an objective, although
transient, division of spacetime into the layers of the growing Past and the diminishing
Future. In particular, thus, it contradicts the theory of Aristotelian possibilism. That, of
course, presents a problem for the transientist. In dealing with it many authors who
declare themselves in favor of transientism pervert their commitment to transientism
into a mere lip service.
Indeed, as it was pointed out above in this essay, according to transientism the
ontological statuses 0/ events - those 0/ being past, present or future, and, as /ar as
116 HELENA EILSTEIN
distinguished in the totality of events. That correlation hardly can be conceived any
other way than by means of relativization ofontological statuses of events.
Indeed, take any fourpoint, P', Located in the spatiotemporal area Op, that is, in the
area outside both lobes of the double light cone of P. Since there is nothing particular
about P, anything that concerns P concerns P' too. It also is the focus of intersection of
some corresponding world-lines, along which, according to our declared transientist,
the Now runs reaching it (when it does) concurrently along all these lines. Instead of
being in a kind of limbo ontological state, P' like P, should thus also 'always' be in one
of the classical states of pastness, nowness or futurity.
Let us thus, consequently, focus on these ontological states of particular fourpoints
comprising Op which are associated with the nowness of P. (Being in one and the same
spacetime as P and being 'always' in definite ontological states these fourpoints cannot
fai! to be in definite ontological states when P is in the state of nowness. Indeed, to the
temporally pointlike ontological state of nowness of P there must correspond adefinite
ontological state of Opas a whole. and this state cannot be anything but the sum total of
the definite ontological states of every particular fourpoint belonging to Op). One
cannot admit that all these fourpoints are in one and the same ontological state, because
it would mean that the nowness of the randomly chosen fourpoint, P, has a cosmic
significance, deterrnining the ontological state of every other fourpoint. Let me add that
the above admission also must result in a contradiction. Indeed, if all fourpoints in Op
were, say, future, then (since there is nothing particular about P). all points outside the
light cone of any of them - say, of one labelIed p'- should also be future, since they
should be future even when p' becomes present That must concern also p. contrary to
our original stipulation concerning the ontological status of P. The assumption that all
these point-components of the outside of the light-cone of P are past when P is present
of course would result in the same kind of contradiction. On the other hand, assurne that
every fourpoint p' in the outside of the light cone of P is present when Pis. Since the
light cones of these outer points intersect with that of P, some fourpoints in spacetime
again would be rendered to be in two ontological states at once. That result thus seems
to call for the abandonment of the transientistic stipulation about the absolute character
of ontological states of events, Only the appropriate relativization of the 'ontological
states' of events to 'ontological states' of other events may lead us out of the
contradiction; and indeed such a way is advocated by some avowed 'transientists', That
relativization, preposterous from the point of view of each version of genuine
transientism, would be particularly preposterous from the point of view of the
possibilistic transientist, for it would force upon us the conclusion that the concepts of
open and extinguished possibility also have a merely relative character.
An alternative against the relativization might be seen in the acceptance of the
conception of a privileged wor/dline. The run of the Now along that worldline would
result in a double-Iobed light-cone gliding along it and unambiguously dividing. at each
of its current positions, the current Past (at and inside the backward lobe) and the
current Future (at and inside the forward lobe) from the area of spacetime comprising
fourpoints to which currently neither pastness nor futurity could be attributed, Special
Relativity, of course, does not know anything about such a privileged worldline.
Neither the nomic order of the universe, nor any kind of global conditions prevailing in
it seem to bestow such a privileged character to any worldline. In order to introduce
118 HELENA EILSTEIN
a kind of legitimate privilege, one presumably would have to assume that there is such a
privileged world-'line' for every particular epistemic subject - and namely, his own
'life-line', to use the expression of Weyl. That, however, would mean subjectivization
of the ontological states of events, contrary to the thesis of authentie transientism.
Transientists, however, can try to meet the challenge offered by the standard
interpretation of Special Relativity in some primafacie more promising ways.
One e.g. may argue that Special Relativity deals with some aspects of time only and for
that reason ignores the reality of the 'flow of time'. Accordingly, one may maintain that
the absolute relation of Copresentness cannot be defined in terms of those absolute or
frame-dependent spatiotemporal and temporal relations which are dealt with in
Relativity.
C.W. Rietdijk exarnined a conception like that in his papers, particularly in 'Special
Relativity and Determinism' . He considers there (of course, without using my
terminology) the idea according to which, besides the frame-dependent stratifications of
the spacetime into instants, there also obtains an absolute stratification of spacetime,
corresponding to the classes of Copresentness of events. Clearly that stratification must
be one into a continuum of curved (in a not specified way) spacelike hypersuTj'aces.
Rietdijk implicitly attributes to the transientist the intuitive assumption that these
hypersurfaces are geometrically identical: the Now consecutively passes through
spacetime as a rigid curved three-dimensional hypersurface. Rietdijk shows that these
presuppositions bring forth absurd consequences: it may happen that an observer (no
matter how long he lives) would never be able to receive a signal from another observer
who is at rest with respect to hirn: the signal sent to hirn would reach his worldline in
the Past.
In her defence of transientism in the paper 'Past, Present, Future and Special
Relativity,53 Nat&a Rakic actually also takes up the idea of the stratification of the
spacetime into the hypersurfaces of Copresentness. She does not mention the above said
paper of Rietdijk and does not explain by means of what additional presuppositions
does she avoid the above disastrous consequences. Neither does she say how the
stipulated stratification of the totality of events into classes of Copresentness might be
established empirically. The lack of that information evidently does not trouble her as
she avows that
the notions of past, present and future are not temporal but ontologicaI notions ...[T]he new
primitive relation [of Copresentness1does not have a place in our physicaI theory.54
In the views of Henry Stapp,55 too, Relativity recognizes only 'Einstein time', that
is, the time in which the already 'fixed' elements of reality, the ones known to
physicists from observations, are located. It does not take into account 'the process
time', also called the 'dynamic' time. That has to be the time in which open possibilities
gradually actualize (or become extinct), in an order which has nothing to do with their
location in the 'Einstein time' .
Stapp's possibilistic transientism considerably differs from the above discussed
version. In his model there is no Now 'running the ladder' of successive hyperspaces in
some privileged stratification of spacetime. One should rather think of each actualized
event 'as a dot in a four-dimensional Seurat-like painting' - a painting whose dots
AGAINST DETENSERS 119
'come into existence in some serial order,.56 Stapp does not make a distinction between
happening and actualization and he points out that
the actuaI is represented not by an advancing, infinitely thin slice trough the spacetime
continuum, but rather by the sequence of actual becomings, each of which refers to a
bounded spacetime region ... We have, therefore, neither becoming in three~imensional
space, nor being in the four~imensional world but rather becoming in the four-
dimensional world.57
It turns out, however, that the successive actualizations (and extinctions) of hitherto
open possibilities in the supposed 'process time' are beyond the scope of our empirical
cognition. 58 As a matter of fact, thus, nothing may induce one to accept this conception
except one's aversion toward permanentism; an aversion reinforced by the belief that
contemporary physics, qua indeterministic, gives a strong support for possibilism.
Some other transientists submit that Special Relativity just does not account for the
actual existence of a privileged stratification of spacetime into instantaneous
hyperspaces. Indeed, according to some physicists Special Relativity in its standard
interpretation is beset with insurmountable paradoxes. Attempts have been made to
formulate an empirically viable theory which would vindicate the concept of a uniquely
privileged stratification of spacetime into instants. With that purpose some authors
propose to abandon both Galilean and Lorentzian formulas of coordinate
transformations in favor of some alternatives.
In his defense of transientism Quentin Smith tries to vindicate the Newtonian
conception of absolute space and absolute simultaneity. (He supports the original
Lorentzian interpretation of the phenomena of length contraction and time dilation).59
He argues for that point of view on a broad conceptual basis. It includes his 'philosophy
of language' (that is, his views on the relation between meaningfulness and truth-values
of sentences on the one hand, and their empirical testability on the other). Its other
component consists in Smith' s view on what constitutes the most fundamental
presupposition of Relativity.60 The third component consists in the author's specific
version of the 'Platonie' theory of abstracts. In Smith's view, abstracts Exist in time, but
not in space (which by itself speils the abandonment of the relativistic concept of
spacetime). The stages of their history have to coincide with the successive Nows; and
the nonspatial character of the abstracts has, according to the argument of Smith, secure
the absolute character to the succession of Nows. Considering the relations between the
respective abstracts and their exemplifications by physical individuals Smith claims to
have established the absoluteness of the relations of simultaneity also between physical
events. 61
Since according to Smith 'physical clocks do not measure the metric of time' ,62 the
relations of absolute simultaneity between events cannot be empirically established.
However, in Smith's view, only supporters of the faulty verificationist 'philosophy of
language' may be induced by that circumstance to question the acceptability of the idea
of absolute simultaneity.
Some other authors who argue for the reintroduction of the Newtonian theory of
space and time into physics do it addressing themselves more directly to difficulties of
contemporary physics.
Physical arguments in favor of reintroduction of the concept of absolute
simultaneity are various. I am not going to mention all of them here. However, the
120 HELENA EILSTEIN
8. THE CONCLUSION.
My above considerations support the claim that permanentism is scientijically
irrefutable. Indeed, they seem to lead to the conclusion that no scientijic, empirically
testable, model of the nomic order of the universe and of its spatiotemporal structure is
AGAINST DETENSERS 121
incompatible with permanentism. Some physical models of the universe may make
permanentism no more plausible than transientism, but there cannot be models which
would make it less plausible than transientism.
That, of course, by no means amounts to the thesis that permanentism is true.
It amounts to no more than the thesis that in case permanentism is false, science would
be unable to discorroborate it. Which, in my view, amounts to the claim that it is
unreasonable for the philosophising animal to commit itself to transientism (instead of,
e.g., leaving the problem unsolved), when it is in the purely cognitive modus of its
psyche. There is no doubt that when it is concerned with the 'timeliness' of its actions,
like jumping overboard an endangered ship or running from a predator, or hitting an
aggressor, or kissing its angry mate, it is bound to manifest its commitment to
transientism.
Helena Eilstein
Korotynskiego 28 m. 91, 02-123 Warsaw, Poland
E-Mail: lena@waw.pdi.net
NOTES
1 Some passages from my paper 'Prof. Shimony and the 'Tmnsient Now'" 1996, Synthese, 107, and from my
book Ufe Contemplative, Ule Praetical,an Essay on Fatalism (in the series Pouum Studies in the
Philosophy 01 the Scienees and the Humanities, 1997, Amsterdam: Rodopi) are incorporated into this text
with slight pamphmses.
2 D. Zeilicovici, 'A [Dislsolution of MCTaggart's Paradox', 1986, Ratio, 28, p.189.
3 The book of S. McCall, A Modelolthe Universe, (1994, Oxford: Clarendon Press) amounts, despite of some
inconsistencies that occur in it, to a very clear exposition and justification (from the author's point of view)
of that variety of possibilism. A more sophisticated presentation and elaboration of the basic ideas of
classical possibilism - a conception to a consideral3le extent inspired by Nuel Belnap (who in bis turn was
much inspired by McCall's viewon time) - is brought in the paper ofT. Placek, 'Bmnching for a Tmnsient
Time' , in this collection. Being not committed to tmnsientism I am not going to analyze the very interesting
comments of Placek conceming the difference between bis views and those of other transientists,
particularly McCall and D. Lewis.
4 With exception ofthe names 'Past', 'Present' (or 'Now') and 'Future', where the use ofcapitalletters marks
the distinction between them and the predicates 'is past', ete., which are used by transientists with reference
to particular events, capitalletters in my text mark the 'tenselessness' of the corresponding expressions, as in
the case of 'Copresentness' above or 'Existence', 'Exists', 'Oeeurrs' ete. below. This convention is
neglected, however, when obeying it is not needed in order to avoid misunderstanding.
S H. Weyl, Philosophy 01 Mathematies and Natural Seience (1946, Princeton Univ. Press), p.l16. According
to my convention conceming the tenseless mode of verbs, I used the capitalletter for 'ls' although it did not
occur in the original Weyl's text.
6 See H.-D. Zeh, The Physical Basis olthe Direetion olTime, (1986, Berlin: Springer Verlag), p.l49. Some
authors claim that Einstein vacillated in bis support of permanentism. This is immaterial in the present
context; however, it is worth noting that the above quotation is laken from a letter Einstein wrote few weeks
before his death.
7 K. Gödel, 'Relativity and ldealistic Philosophy', in: P. A. Scbilpp (ed), Albert Einstein Philosopher-
Seientist; The Ubrary 01 Uving Philosophers, 1951, New York: Tudor Pub!. Co., p. 561. Gödel was
convinced that the antecedent of the above conditional is true.
8 J. Earman, Bangs, Crunehes, Whimpers and Shrieks, 1995, iliford Univ. Press, p. 97.
9 Realizing the danger of confusion. some authors like N. Oaldander graphically distinguish 'Tense'
in the meaning oftempoml tmnsience from 'tense' in its ordinary gmmmatical meaning.
10 Note that in the above combination the symbol T, as associated with the !enn 'tenser' has another
11 LN.Oaklander, 'A Defense of the New Tenseless Theory of TIme' in: LN. Oaklander and Q.Smith (editors
and authors of a number of entries), The New Theory 0/ Time, 1994. New Haven and London:
The University Press, p. 57.
12 To be sure, the information may be adequate or inadequate - true or false. The point is that the
corresponding tenseless sentence should, in the views of the detenser, capture the total information
genuinely penaining to the objective reality, cleansing it from the marks of the egocentric attitude of
the author of the original sentence token.
13 By the purely cognitive discourse (including one's discursive thinking) I mean the discourse whose aim
consist in attaining a possibly adequate account of the objective reaIity, and not also in expressing people's
feelings or assisting people in achieving these or other practical aims. In the words of N. Oaklander, 'given
that we are representing reality, a tensed Ianguage is eliminable in terms of a tenseless one, even though a
tensed language cannot be translated in terms of a tenseless one'. - L N. Oaklander, 'Freedom and the New
theory ofTime', in: R. Le Poidevin, (ed.), Questions o/Time anti Tense, 1998, Oxford: Clarendon Press,
p.187.
14 It should be noted that not a11 indexicals have a subjective character. '/' is obviously subjective, and so
is 'you [the intended addressee of my message]', but not e.g. 'he', whose meaning at a given occurrence
depends on who has been spoken about in the preceding context; not 'yes' whose meaning at a given
occurrence depends on the content of the preceding question. In the views 0/ the transientist such
spatial indexicals as 'here' and 'over there' obviously are subjective, but there is nothing subjective in
the basic temporal indexicals Iike 'now', 'in the Past', 'in the Future', as weil as in the tensed
expressions whose meanings depend on the above.
15 Q. Smith, 'The Phenomenology of A-Time', in the above mentioned collection of LN. Oaklander and Q.
Smith, pp. 351-352.
I. L N. Oaklander, 'Freedom and the New Theory of Time' , in the above named collection of R. Le
Poidevin, p. 197.
11 That presumably concems also communities whose native languages do not provide for tensed forms
ofverbs. There are other means than tenses to manifest the transientistic character of one's world-outlook.
18 L N. Oaklander, Temporal Relations anti Temporal Becoming, 1984, Lanham: University Press
of America, , p.l.
19 Q. Smith, 'General Introduction' in The New Theory o/1ime, op. cil., pp. 4-5.
20 I do not claim that determinism as such should be branded as a prejudice. It is an ontological
hypothesis, and its fate is not obviously doomed in the light of contemporary science. It is only its
aprioristic and undisputed character on the basis of c1assical science that conferred to it the character of
a prejudice.
21 Tbe procedure of assigning synonyms has been ridiculed by A. Prior in his influential paper 'Tbank
Goodness That's Over', where he points out that one's sight of relief, 'thank goodness it is over', issued
e.g. at the conc1usion of a painful session in a dentist's chair, 'certainly does not mean the same as e.g.
'Tbank goodness that the date ofthe conc1usion ofthat thing is Friday, June 15, 1954', even ifit be said
then. 'Neither, for that matter does it mean ''1bank goodness that the conclusion of that thing is
contemporaneous with this utterance. Why should anyone thank goodness for that?'.. , 1959, Philosophy
34,p.17.
22 D.H. Mellor, 'The Need for Tense', in the collection of Q. Smith and LN. Oaklander, p. 24. (Mellor's
contribution is an excerpt from his Real Time, 1981, Cambridge University Press). In the above quotation I
substituted 'Occurs' for 'occurs' in the original text
23 p. 25.
24 G. Nerlich, 'Time as Spacetime', in the collection of Le Poidevin, p. 123.
25 In the paper 'The Past, Present and Future of the Debate about Tense', in his above quoted collection,
p. 29, R. Le Poidevin distinguishes two varieties of the above discussed basic method of reduction.
One, 'the token - reflexive theory' originally was proposed by D.H. Mellor; the other, 'the date theory'
is credited to J.J. Smart. According to the first one, any token u of 'e is occurring now' is true if and
only if u is simultaneous with e. According to the second one - the one evidently nowadays endorsed
also by Mellor, any token of 'e is occurring now' tokened at t is true if and only if e Occurs at t. It is
clear that these proposals are equivalent and my above criticism applies to both of them.
26 J, Earman and R. Gale, the entry 'Time' in R. Audi (ed.), The Cambridge Dictionary 0/ Philosophy, 1995,
Cambridge University Press, p. 804.
21 Ibid. It is a mystery for me what particular help in his elfort to make his concepts clear might the transientist
gain from the stipulation of the existence of 'Mr. X'. X's actions would have to be described in a tensed
AGAINST DETENSERS 123
language, since he would be supposed to successively make the instantaneous components of "the Shift" to
happen. Thus any sentence conveying infonnation about his role could be considered meaningful only if the
specific expressions of the tensed language are meaningful. If our authors allege that there is an inextricable
logical link between transientism and theism, then they failed to make their point.
28 See e.g. the contribution of D.H. Mellor to New Theory of Time, op. cit., pp. 30-31. L. N. Oaklander
advocates the same position, e.g. in 'A Defense of the New Tenseless Theory of Time', in the same
collection and in his above quoted paper in the collection of R. Le Poidevin.
29 Life Contemplative, Life Practical.
30 Terms like 'striving', 'attempting', 'pursuit' belong to the possibilistic conceptual frame. According to the
permanentist they thus also characterize nothing but intensions of some mental states of human individuals.
31 See e.g. his paper 'The Unreality and Indeterminacy ofthe Future in the Light ofContemporary Physics' in:
D. R. Griffin (ed.), Physics anti the Ultimate Significance ofTime, 1986, New YQrk: State University of
New York Press, pp.301-304.
32 Op. cit, p. 804. One encounters the argument in question also with other detensers.
331bid.
34 This stipulation is contained e.g. in C. D. Broad's term 'absolute becoming '.
35 Two papers (by permanentists) in the collection of R. Le Poidevin, viz. J. Butterfield's 'Seeing the
Present' and G. Nerlich's 'Time as Spacetime' bring interesting remarks about how some features of
our everyday, macroscopic, perception of the world foster transientistm. (Some of their claims
concerning the character of human perceptions are submitted to criticism in R. Teichman's review on
that collection in The British Journal for the Phil. of Sei., December 1999).
36 _ which, in contradistinction to a process, should not consist in achain of events - that is, in a chain
of its causally bound stages.
37 See his book, The Natural Philosophy ofTime, 1980, Oxford: Clarendon Press.
38 See his book, Bergson and Modem Physics, 1971, Dordrecht, Holland: Reidel, and his paper 'Two
types of continuity' in the collection of R.S. Cohen and M.W. Wartofsky, Logical and Epistemological
Studies in Contemporary Physics, 1974, Reidel.
39 M. Heller, 'Granice przestrzeni i czasu [The Limits of Space and Time)' in: M. Heller, J. M!\Czak and J.
Urbaniec (eds), Granice nauki [The Limits ofSeienceJ, 1997, Tam6w: Biblos Publ. House, p.73.
40 C.D. Broad wanted to avoid it by means of introducing the concept of 'absolute becoming'.
41 The most fundamental concept of A. N. Whitehead's theory of time (wh ich I do not discuss here
because of its slight impact on the contemporary dispute between transientism and permanentism) is
that of the atomicity of actual occasions which are supposed to enter the physical time as indivisible
wholes. As it is indicated by S. Rosenthai, in developing his theory 'Whitehead was motivated [among
other needs] by the need to avoid Bergson's seamless duration'. ( See S. B. Rosenthai, Time, Continuity
and Indeterminacy; a Pragmatic Engagement with Contemporary Perspectives; 2000, SUNY, p. 20).
42 Determinism is compatible with, but does not presuppose or imply the efficacy theory of causation.
43 lf the volume of the spatial area in question could be arbitrarily smaIl, we might irnagine that it cou1d be
contracted into a point. In that case transspatial determinism would say that every point in space is an aleph,
in the sense ofthe J. L. Borges' fantastic story 'Aleph' (except that looking at an aleph one immediately saw
everything that was currently going on in the universe).
44 C. Rovelli, 'Analysis ofthe Distinct Meanings ofthe Nation ofTime in Different Physical Theories', 1995, 1I
Nuovo Cimento, Vol. 110 B, N. I, pp. 81 and 86.
45 M. Heller and Wieslaw Sasin, 'Emergence ofTime',1998, Physics Letters A, 250, p.48.
46 S. Hawking, The Edge of Spacetime', in: P. Davies (Ed.), The New Physics, 1992, Cambridge University
Press,p.68.
47 Zygon, September 2000, vol. 35, no. 3, p. 674.
48 Op. eit. (1997, Allen Lane, The Penguin Press, p. 283-285.
49 See e.g.: J. Eannan, 'Recent Work on Time Travel', in: S.F. Savitt (ed.), Time's Arrows Today, 1995.
Cambridge University Press. 'I do not see - Eannan also points out in that work (p. 310) - any prospect
for proving that time travel is impossible in any interesting sense [evidently, when such questions as possib1y
prohibitive costs are left aside - H. E.] ... [N]o proof of this impossibility has emerged in classical GRT'.
The author admits however, that '[s]tandard quantum mechanics is hard to reconcile with CTCs'.
In Scientific American, Jan. 2000, L. H. Ford and T. A. Roman discuss the problem of the possibility of time
travel in another perspective, namely in connection with the concept of negative energy which, as they point
out, must be incorporated in contemporary physics. The upshot of their considerations is utterly sceptical,
which, of course, does not rule out the other approaches.
124 HELENA EILSTEIN
50 See Olivier Costa de Beauregard, Time, the Physical Magnitude 1987, Dordrecht: Reidel Publishing House,
p.252.
51 Compare A. Shimony, 'Tbe Transient Now', in: A. Shimony, 1993, Searchfor the Naturalistic World-View,
Cambridge University Press, p. 275.
" See: H. Putnam, 'Time and Physical Geometry', in his Philosophical Papers, 1975, vol. I" Cambridge
University Press. C. W. Rietdijk, 'Special Relativity and Determinism', 1976, Philosophy of Seience,
43. Unfortunately Rietdijk employs the term 'determinism' not only for determinism but also for
actualism. Tbis flaw, however, does not detract from the soundness of his reasoning.
53 1997, The British Journal for the Phil. of Sei., volume 48. Nr. 2.
54 L C., p. 275 and 276.
55 See his 'Einstein Time and Process Time', and the discussion of that paper by a number of authors in: D. R.
Griffin (ed.), Physics and the Ultima te Significance ofTime, 1986, State University of New York Press.
56 See the comments ofW.B. Jones on the theory ofStapp, p. 281 in the above collection.
57 Op. eit., p.267.
58 Stapp acknowledges that; see op. cit., p. 264.
59 See his Language and Time, 1993, Oxford University Press, and particularly his paper, 'Absolue
Simultaneity and the Infinity ofTime' in the collection of R. Le Poidevin.
60 Tbat interpretation of Relativity is submitted to criticism in the paper of G. Nerlich 'Time as
Spacetime' , in the above collection, where the author critically deals also with other views of Smith' s.
61 Since according to Smith's Platonism the Existence of abstracts does not require that they possess
exemplifications at all times, the author also argues in favor of the view that, regardless of whether
there are, or not, reasons to accept the hypotheses of the Bing Bang and tor the Big Crunch, 'past and
future time are infinite' and 'time as a whole consists of an infinite number of infinitely long temporal
series'. (P. 136).
62 'Absolute Simultaneity .. .', I. c., p. 136.
6J Physics Today, parts land 11, March and April 1998.
64 Part H, p. 42.
65 See e.g. the papers of Simon Saunders, 'Tbe Quantum Mechanics and Tense', 1996, Synthese, 107, and
of M. Lookwood, 'As Time Goes By', International Studies in the Philosophy of science, vo!. 1 I, n.l,
1997.
66 Op. eit., p. 22, p.29.
REFERENCES
Belnap, N. 'Branching Space-time', 1992, Synthese, 92.
Broad, c., 'Ostensible TemporaIity', in the collection ofR. Gale.
Beauregard de, 0., Time, the Physical Magnitude 1987, Dordrecht, Holland: Reidel Publishing House.
Butterfieid, J., 'Seeing the Present' in the collection of Le Poidevin.
Capek, M., Bergson and Modern Physics, 1971, Dordrecht, Holland: Reide!.
-'Two Types of Continuity', in the collection of R.S. Cohen and M.W. Wartofsky, Logical and
Epistemological Studies in Contemporary Physics, 1974, Reide!.
- 'Tbe UnreaIity and Indeterminacy of the Future in the light of Contemporary Physics' in the collection of
D. Griffin.
Davies, P. (ed.), The New Physics, 1992, Cambridge University Press.
Deutsch, D., The FabricofReality, 1997, London: Tbe Penguin Press, Allen Lane.
Earman, J., Bangs, Crunches, Whimpers and Shrieks, 1995, Oxford Univ. Press.
-'Recent Work on Time Travel', in the collection of S. Savitt.
Earman, J. and Gale, R., 'Time' in: R. Audi (ed.), The Cambridge Dictionary of Philosophy, 1995, Cambridge
University Press.
Eilstein, H., 'Prof. Shimony and the 'Transient Now'" 1996, Synthese, 107.
- Life Contemplative, Life Practical, an Essay on Fatalism. in the series Poznan Studies in the Philosophy of
the Seiences and the Humanities, 1997, Amsterdam: Rodopi.
Ford, L., Roman, T., 'Negative Energy, Worm Holes and Warp Drive', Scientific American, Jan. 2000.
Gale, R.M., The Language ofTime, 1968, London: Routledge and Kegan Pau!.
-{ed.), The Philosophy ofTime, 1978, New Jersey: Humanities Press.
Griffin, D.R. (ed.), Physics and the Ultimate Significance ofTime, 1986, State University ofNew York Press.
AGAINST DETENSERS 125
Goldstein, S., 'Quantum Theory without ObservelS', Physics Today, March and April, 1998.
Gödel, K., 'Relativity and Idealistic Philosophy', in: P. A. Schilpp (ed), Albert Einstein Philosopher-Scientist;
The Library of Living Philosophers, 1951, New Yorl<:: Tudar Publ. Co.
Hawking, S.W., 'The Edge of Spacetime', in the collectoin of P. Davies.
Heller, M., 'Granice przestrzeni i czasu [Tbe Umits of Space and Time]' in the collection of M. Heller,
J. MllCzak and J. Urbaniec.
Heller, M., Ml\czak J., Urbaniec, J.,(eds), Granice nauki [The Limits of ScienceJ, 1997, Tam6w: Biblos
Publ. House.
Heller, M., Sasin,W., 'Emergence ofTime' ,1998, Physics Leners A, 250.
-'Time and Physics - a Noncommutative Evolution', this collection.
- 'Cosmological Singularity and the Creation of the UnivelSe', Zygon, September 2000, vol. 35, no. 3
Jones, W.B., 'Physics and Metaphysics: Henry Stapp on Time',in the collection ofD.R. Griffin.
L.e Poidevin, R. 'The Past, Present and Future of the Debate about Tense', in his collection.
L.e Poidevin, R. (ed.), Questions ofTime and Tense, 1998, Oxford: Clarendon Press.
Lockwood, M., 1997, 'As Time Goes By',lntemational Studies in the Philosophy ofscience, vol. 11.
McCall, S., A Model ofthe Universe, 1994, Oxford: Clarendon Press.
MeTaggart, J.M.E., 'Time', in he collection ofR. Gale.
Mellor, D.H., 'The Need for Tense', in the collection of Q. Smith and L.N. Oaklander.
Nerlich,G., 'Time as Spacetime', in the collection ofR. L.e Poidevin.
Oaklander, L.N., 'A Defense of the New Tenseless Theory of Time' in the collection of L.N. Oaklander and Q.
Smith.
- 'Freedom and the New Theory ofTime' , in the collection of R. L.e Poidevin.
- Temporal Relations and Temporal Becoming, 1984, Lanham: UnivelSity Press of America.
Oaklander, L.N., Smith, Q. (eds.), The New Theory ofTime, 1994, New Haven and London: The UnivelSity
Press.
Placek, T., 'Branching far a Transient Time' , this collection.
Prior,A., 'Tbank Goodness That's Over', 1959, Philosophy 34.
Putnam, H., 'Time and Physical Geometry', in his Philosophical Papers, vol.I, 1975, Cambridge
UnivelSity Press.
Rakit, N., 'Past, Present, Future, and Special Relativity', 1997, The British Journal for the Philosophy of
Science, volume 48. Nr. 2.
Rietdijk, C.W., 'Special Relativity and Determinism' , 1976, Philosophy ofScience, 43.
Rosenthai, S. Time, Continuity and Indeterminacy; A pragmatic Engagement with Contemporary
Perspectives, 2000, SUNY.
Rovelli, C., 'Analysis ofthe Distinct Meanings ofthe Notion ofTime in Different Physical Theories' , I 995,
/l Nuovo Cimento, Vol. 110 B, N. I, pp. 81 and 86.
Saunders, S., 'Tbe Quantum Mechanies and Tense', 1996, Synthese, 107.
Savitt, S.F., (ed.), Time's Arrows Today, 1995, Cambridge UnivelSity Press.
Shimony, A., 'The Transient Now', in: A. Shimony, Searchfor the Naturalistic World- View, 1993, Cambridge
UnivelSity Press.
Smith, Q., 'Absolue Simultaneit and the Infinity ofTime' in the collection ofR. L.e Poidevin.
- 'General Introduction', in the collection of L.N. Oaklander and Q. Smith.
- Language and Time, 1993, Oxford University Press,
- 'The Phenomenology of A-Time', in the collection of L.N. Oaldander and Q. Smith.
Stapp, H., 'Einstein Time and Process Time' in the collection ofD.R. Griffin.
Teichman, R., a review of the collection of L.e Poidevin, The Britsh Journal for the Philosophy of Science,
December 1999.
Whib'Ow, GJ., The Natural Philosophy ofTime, 1980, Oxford: Clarendon Press.
Weyl, H., Philosophy ofMathematics and Natural Science, 1946, Princeton UnivelSity Press.
Zeh, H.-D., The Physical Basis ofthe Direction ofTime, 1986, Berlin: Springer Verlag.
Zeilicovici, D. 'A [Dislsolution ofMCTaggart's Paradox', 1986, Ratio, 28.
STEFAN SNIHUR
Considering this question one can come to the conclusion that neither the positive
nor the negative answer to it is satisfactory, because arguments for the nonexistence
of the future seem equally well-founded as arguments for its existence. To recall
some of these arguments, let us consecutively focus on each of the two contradictory
theses concerning the existence of the future.
The main reason why the truth of the above thesis seems dubious is based on the
fact that any domain of time, such as the future, may comprise only objects which
belong to the widely understood sphere of material beings; however no future
objects could meet the empirical criterion of existence (ECE for short) which is
considered legitimate just with respect to beings of that category. According to that
criterion existence may be granted to such objects only which can directly or
indirectly manifest their existence in these or other kinds of sense experience.
To put it simply, in the light of ECE whatever exists is either an object of some
actual direct or indirect observation or something that could be observed if
a physically possible observation was performed at an appropriate time and place.
But from the point of view of both common sense and science the very concept of
observation rules out the possibility of either directly or indirectly observing
anything which in the order of time would follow the respective act of observation.
In other words, in view of ECE no future objects from the widely understood sphere
of material beings can be granted existence. On the other hand, as it has been
127
H. Eilstein (ed.), A Collection of Polish Works on Philosophical Problems
ofTime and Spacetime, 127-136.
© 2002 Kluwer Academic Publishers.
128 STEFAN SNIHUR
pointed out above, no objects from beyond that sphere can have a location in time.
Obviously thus neither of them can ever be considered future. That leads to the
conclusion that there are no future beings - and, consequently, that the future does
not exist.
need to resort to that reason, because our own 'dwelling' in the present and our usual
awareness of that fact provides us with a sufficient confinnation of that thesis.
It seems also right to maintain that our own present state together with everything
that is simultaneous with it constitute for us the sphere of beings which enjoy the
highest degree of realness. All other existing beings seem to be endowed with
a merely derivative kind of reality, based on what exists just now.
It seems thus right to attribute to the present objects a unique mode of existence,
namely the real, or, in other words, actual existence.
To the second class of objects, whose existence one can acknowledge in virtue of
ECE, belong those which satisfy just the condition (ii). They are usually called
'past objects'. Thus ECE also confirms the thesis of existence of the past.
There are many other arguments in support of the last thesis.
Firstly, its rejection brings forth some essential complications as far as
understanding the time and the definition of truth-values of assertions about things
past are concerned. (The last difficulty is analogous to that one which was already
mentioned in the preparatory examination of the negative thesis concerning the
existence of the future).
The point of departure of the second kind of arguments consists in that it is
possible to meaningfully think and talk about some particular past objects, events
and states of affairs as weIl as about the past in general. In view of the general
assumption that it is impossible to meaningfully speak about things to which in
principle no kind of existence may be granted one has to attribute a kind of
existence to the past.
For the third group of arguments the point of departure consists in existence of
creatures endowed with consciousness. A conscious being dweIls in the world not
just as its passive constituent subject to the laws of nature. Its main feature consists
in striving to preserve its individuality. Its existence is thus inseparably associated
with its involvement in the creation of its specific inner reality and, in particular,
with the development of its more or less clear awareness of itself as an individual
conscious being. This creative process, however, would be impossible, if the past
amounted to an absolute non-being - and thus if no past things were represented in
one's memory (since one cannot remember 'nothing'). For indeed, among various
elements constituting one's consciousness the most important and indispensable
ones are those which link it with the past, and through the past lead it to the present
and the future. This allows us to state that existence of conscious beings is possible
just because their past states do not suffer an absolute annihilation but, although no
more real, exist in a certain way.
Since, however, every conscious being is a kind of microcosm, one can admit, by
virtue of analogy, that in the wholeness of being the loss of reality by particular
objects is not identical with their absolute annihilation. Whatever ceases to exist as
a present object, remains in being as an element of the domain we define as the past.
On the other hand it follows from the above, that while we acknowledge that the
past exist in a certain way, we have to deny it the attribute of reality. That is worth
emphasizing, because of the inclination of human mind to identify all modes of
existence with the real, present existence, that is, with the mode of existence of
things, which may be 'given' to us in the most conspicuous way. That inclination
EXISTENCE OF THE FuTURE 131
makes it difficult for us to acknowledge existence of past and future objects as weil
as of those which belong to immaterial domains of being. In order, thus, to both
highlight the difference in modes of existence between the past and the present and
at the same time point out the ontological links between these domains, the name of
'post-real existence' is assigned here to the mode of existence of objects of the past.
It is thus seen that in view of the above considerations it turns out to be
necessary to introduce the distinction between the concept of existing in the absolute
sense and the concepts of this or that particular mode of existence.
The concepts of real and post-real existence make it possible to put forward
a new version of the negative thesis conceming existence of the future:
(B') The future does not exist in either the real or the post-real way.
This thesis allows one to admit existence of the future provided it also is
admitted that the mode of existence of future things differs both from the real and
the post-real existence.
In support of (B'), one can point out the previously indicated fact, that objects
intentionally located in the future do not meet ECE. On the other hand, in view of
the above made distinction a new question arises: Is it indeed possible to grant
a mode of existence to future objects, and thus to the future?
I believe my above considerations justify the positive answer to this question.
As a matter of fact all arguments for existence of the past, with the obvious
exception of the one appealing to ECE, can be reformulated so as to become
arguments for existence of the future. That means, however, that one has to inquire,
in what way does the future exist?
At this point of our considerations a crucial importance has to be attributed to the
idea that all and only such objects which can be legitimately considered future 'may'
become real in the course of time. Accordingly, it seems appropriate to assign the
name of potential ( pre-real, possible) existence to that mode of existence which is
specific for future objects.
Let me stress that the concept of possibility (potentiality) is understood here in
a specific, temporal way. It is different than when one speaks e.g. of logical
possibility or of physical possibility of occurrence of things of a given kind in these
or other spatiotemporal locations. In the present context we speak of potentiality to
become real, a potentiality which is specific for such particular objects whose
present non-reality is associated, in the prevailing situation, with their 'disposition'
to become real..
With the help of that concept of potentiality one can present the positive thesis
concerning the existence of the future in the following way:
(A') The future exists potentially.
In contradistinction to (A) and (B), the theses (A') and (B') do not contradict one
another. Together they amount to a general answer to the question concerning
existence of the future. This answer does not, however, characterize the ontological
status of the future in a sufficiently precise way. It is thus now necessary to consider
more closely the other problem named in the abstract, namely: what is the future?
132 STEFAN SNIHUR
(Tl) The future is the domain of objects which exist either potentially or quasi-
potentiall y.
(T2) The future is the domain of objects which exist potentially.
(T3) The future is the domain of objects which exist quasi-potentially.
It is obvious that (T3) cannot be accepted. In its view the future would comprise
such objects only which will never become real. That would be in a crass
contradiction with the fact that every presently existing object as weil as every
object of the past once belonged to the future.
It is more difficult to make a choice between the other two theses.
(Tl) seems to agree with the way we usually conceive the future. For indeed,
when we think and speak about the future, we take it into account that some
possibilities will and some will fail to actualize. And since we never are absolutely
certain which objects will become real, it appears unavoidable to include objects of
both the above categories in the range of the future.
In philosophy of time one encounters the idea of the future as the domain of
possibilities which in many cases 'compete' with one another for actualization.
That idea is expressed in the claim that, in contradistinction to the past and the
present, the future is 'open'. The common sense intuitions incorporated in (Tl) agree
with this view. In favor of (Tl) also seem to speak those arguments which are often
leveled in philosophy against (T2). The laUer thesis reduces the rich conception of
the future as consisting of many alternative variants to the idea of the future as
represented by only one variant.
(T2) finds its fairly accurate expression in the claim that the future is 'what is
going to be'. However, 'what is going to be' is often identified with 'what is bound
to happen'. It is therefore easy to conclude that (T2) is unacceptable, because it
deprives the future of its specific openness and characterizes it in a way which is
typical for strict determinism and fatalism. As a consequence, (T2) seems
EXISTENCE OF THE FUTURE 133
incompatible with the belief that humans are endowed with free will, in the sense in
which being endowed with free will means being able to consciously influence the
process of actualization of some potential states of affairs.
Despite appearances, the above claims do not refute (T2) and provide a decisive
support for (Tl). Indeed, let us first of all note that they are based on the illegitimate
assumption that according to (T2) every 'entrance' of some potential objects into the
real sphere of being must represent the manifestation of an inevitable necessity.
This assumption is unacceptable, because potential objects, in the sense which is
presupposed here, are just those which de facto will become real, but not necessarily
such ones wh ich are bound to become real. The reference to the potential status of
some objects and states of things does not by itself legitimize any particular ans wer
to the question, why these and not other objects and states of things will become
real. Consequently although (T2) corresponds to the point of view to which the name
of 'logical determinism' is often assigned, it does not provide any support for the
doctrines of strict deterrninism, fatalism, for the refutation of the idea of free will
etc.
In general, both (Tl) and (T2) leave the important and complicated problem of
determinism and free will open. It thus will be omitted in our further discourse
concerning the question, which of these two theses represents the adequate
description of the future.
Now, despite the criticism of the above mentioned arguments leveled against
(T2), it may seem that (Tl) is the one that should be accepted
Two reasons in favor of this view are worth to be pointed out.
Firstly, there is no doubt that (Tl) and not (T2) is close to common intuitions
concerning the future. Secondly, it may seem that (Tl), in contradistinction to (T2),
provides the way to more fully disclose the ontological specificity of the future, its
complex inner 'structure', its ties with the other temporal domains of being.
The acceptance of (Tl), however, brings forth some serious difficulties in the
analysis of a number of problems of time, such as the problems pertaining to the
temporal succession of things and to the character of the temporal continuum. These
difficulties mainly result from the fact that according to (Tl) the future is
ontologically heterogeneous, since it is supposed to comprise both potential objects
and those of a rather obscure ontological status, namely the putative
'quasipotentially existing ones'.
On the basis of (T2) this heterogeneity and the difficulties associated with it are
elirninated. Accordingly, (T2) offers a simpler, more convenient point of departure
for the analysis of some important problems of time. At least prima facie that
seems to provide a sufficient reason for the acceptance of (T2).
In view of the above remarks the task of making a choice between (Tl) and (T2)
evidently requires to have at one's disposal some appropriate criterion. In order to
find such a criterion we should address ourselves to a problem, which would be both
of considerable philosophical importance and associated with basic questions
pertaining to the ontological status of the future. The above requirement is met by
the problem, whether the fundamental logical principles - the laws of excluded
rniddle and of non-contradiction - are valid for assertions concerning the
components of the future.
134 STEFAN SNIHUR
That is a problem which for a long time has drawn attention of philosophers, like
Aristotle, Kotarbiitski, Lukasiewicz and many others. It is impossible to discuss it
here in all its complexity. I will restrict myself to a consideration of one of its
important partial aspects. That is the question: what ontological assumptions about
the future have to be accepted in order to justify the claim that the above logical
principles are valid with respect to statements conceming future states of affairs. But
this lirnited task, too, cannot be undertaken without making some strong preliminary
assumptions. They can be presented in the following form:
(H) The principles of non-contradiction and excluded rniddle can be valid for all
sentences of a given language J only if all the following conditions are met:
(1) There are only two truth-values: Truth and Falsehood.
(2) Every sentence of J except those which contain temporal occasional
expressions is permanently possessed of just one of the above truth-values.
(3) The sentential connectives 'it is not the case that', 'and' and 'or' are
endowed with their standard meaning.
(4) Two sentences which do not contain temporal occasionals are never both
true or both false in case they express contradictory propositions (Z and not-Z).
(H1 Those sentences of the said language which contain temporal occasional
expressions (like 'now', or 'future' or tensed forms of verbs) must meet the
following conditions which have to provide for a modified version of the above
fundamental principles of logic:
(1 ') As above in (1)
(2') Each of such sentences is possessed at each instant of time (though not
permanently) of just one of the above truth values.
(3') As above in (3).
(4') At every instant of time no two sentences which express contradictory
propositions are both true or both false.
In particular, thus, the condition (2) has to be satisfied for the laws of non-
contradiction and excluded middle to be valid with reference to all such sentences
of the language in question, which do not contain temporal occasional expressions.
Generally, the satisfaction of the above conditions depends on two factors.
Firstly, it depends on the syntactic and semantic specificity of a given language: on
the nature of its terms and the character of the ties between the concepts expressed
by these terms and the presupposed object domain (the universe of discourse) of
that language. Thus, the conditions in question are not satisfied with respect to a
language, which contains vague terms, nontemporal occasional expressions etc. The
second factor consists in the ontological specificity of the presupposed object
domain (the object domain, for short) of the respective language. In particular, as far
as the second factor is concemed, it is necessary to distinguish two kinds of
languages: the temporal and the nontemporal ones. A characteristic feature
of a language of the first kind consists in occurrence, within its object domain,
of objects, which exist in time, and thus come into being, undergo various changes
and subsequently lose their reality. That is the domain of objects, which belong to
EXISTENCE OF THE FuTURE 135
the instant m to formulate in J at least one sentence Z that would be free of temporal
occasionals and predicate something true of a. However, in view of the general
principle that it is impossible to truly predicate about nonexisting things one has to
admit that at the instant m' Z cannot be true. Consequently J would fail to satisfy
the above named principles of logic.
The presented considerations, I believe, justify the thesis that the following
condition must be satisfied for the validity of the principles of non-contradiction
and excluded rniddle in temporallanguages:
(*) Every object belonging to the temporal domain of being exists in the
absolute sense at every instant of time.
The condition (*) points out to a specific negative criterion of choice between
different ontological suppositions conceming the past, the present and the future. In
view of that criterion one should reject the suppositions which are incompatible with
the satisfaction of (*).
In particular, as far as the character of the future is concemed, the above criterion
seems to allow us to resolve the question of the adequate choice between (Tl) and
(12). It seems, indeed, to be the case that on the basis of (Tl) the condition (*)
cannot be satisfied, because quasi-potential objects would cease to exists in any
sense when their supposed temporal location no more belongs to the future. For
example, the day before yesterday, that is, on the 31 s1 of May, 1996, I could
justifiably state that there is a possibility of an eruption of Vesuvius on the 1st of
June 1996. I cannot maintain that this possibility still obtains now. That makes it
questionable whether the possibility of an eruption of Vesuvius was ever a genuine
component of the future.
(12), in contradistinction to (Tl), does not leave place for any doubt conceming
its compatibility with (*). In view of this, one can assume that the adequate ans wer
to the question: what is future? is provided by (12). The future comprises only such
objects which will become real in the course of time.
Ste/an Snihur
Krasinskiego 18 m.49, 01-581 Warsaw, Poland
REFERENCES
Aristotle. Hermeneutics.
Kotarbinski, T .• Zagadnienie istnienia przyszloSci [The Problem of Existence of the Futurel. Przeglqd
Filozojiczny XVI, 1913.
I:..ukasiewicz. J. 1987: 0 zasadzie sprzecznosci u Arystotelesa {On the Principle of Contradiction in
Aristotlel. PWN. Warszawa.
ANDRZEJ POLTAWSKI
Abstract. Roman Ingarden was an outstanding disciple of Edmund Husserl's. His conception of time
grew in the context of his endeavour to solve the realism-idealism issue. The crucial text is 'Man and
Time', initia11y his lecture at the IX International Congress of Philosophy in 1937. While analysing two
ideas of time - regarding as existent but the content of the actual moment or, on the other hand,
acknowledging the existence of enduring in time, real and acting human persons, Ingarden embraces the
second conception, showing the aporiai to which the first conception leads. Consequently, the basic
meaning of 'constitution' (see Sect. 5) is for hirn the development of a living creature or of a human
person and not, as for Husserl, the creation of sense in the flux of consciousness. Nevertheless, his
embrace, for epistemological reasons, of the concept of 'pure conseiousness' - in spite of his doubts
coneerning a11 the ontie features attributed to it by Husserl - makes his idea of time ambiguous.
The first text in which Ingarden tries to formulate the issues pertaining to time is his
lecture at the IX International Congress of Philosophy (Congres Descartes)
in 1937. 1
There are - he says - two fundamentally different ways of experiencing time and
ourselves in time, mutually excluding one another and seemingly equally valid. The
137
H. Eilstein (ed.), A Collection 01 Polish Works on Philosophical Problems
01 Time and Spacetime, 137-148.
© 2002 Kluwer Academic Publishers.
138 ANDRZEJ PÖLTAWSKI
extreme polarity of these experiences makes them the ultimate basis of mutually
opposed metaphysical standpoints in the history of European philosophy, from the
controversy between Heraclitus and the Eleatics till the contemporary conflict
between realism and transcendental idealism. Yet the difference between the two
experiences of time seems to be most acutely reflected in the problem of the essence
of man and makes this essence into the central problem of philosophy.
The problem belongs, then, to metaphysics. But none of the two positions is,
according to Ingarden, a quite arbitrary construction or amistake, because there
exist concrete experiences which correspond to each of them. Thus, the solution to
the controversy idealism-realism implies adecision concerning the validity of one of
these experiences, and this decision is tantamount to choosing one of the two
diametrically different ideas of man and of the world in which man lives.
Ingarden's description of the first experience of time -let us say the realist one
- characteristically stresses the role of real action and, consequently, the necessity
of one's self-identity, one's being one individual, and of remaining the same human
being by the experiencing and acting person in the lapse of time. This idea of man
transcends the realm of pure consciousness which, in this experience, is given as but
a symptom of realforces.
The other - let us call it 'idealist' way of experiencing time has - says Ingarden
- its source in CI) my 'becoming aware of the destructive role that time has for my
existence' and (2) in my 'arriving at the conviction that as a person I am myself
constituted only in multifarious temporal perspectives of experiences'? In this
attitude, it seems that what really exists is but the actual, punctual moment; thus,
everything, including myself, would have to constitute itself in the flow of such
punctual moments. The character of the present is in this experience radically
different than in the first one, where the present was given as a concrete phase of my
history. In fact, the very real existence of myself as a human person is here being
denied.
At the end of the Paris lecture Ingarden concludes that we must seek new
solutions to the problem of the essence of the human being, the problem formulated
by Descartes in the words 'Nondum vero satis intelligo, quis nam sim ego ille, qui
iam necessario sum'.
The Polish text of the lecture, published in 1938, includes a third part which was
not included in the Paris lecture. Ingarden writes:
... a contemplation of how my 'I' is constituted for me indeed strongly undermines my
belief that lexist as a human being who transcends his actual experiences and his actual
present. But at the same time this contemplation is not capable of banishing this belief
entirely[ ... ]. Thus, aB sources of knowledge about oneself need to be scrupulously
examined, and we need to consider whether and within what limits the results of this
knowledge are capable of assuring us about the existence and the properties of ourselves
as people transcending experience. It is a problem for the solution of which we are as
yet unprepared. For no matter how much effort has been expendet in the his tory of
European epistemology upon the investigation of the process of sensory perception and
of the cognitive results acquired on its basis, still, by strange coincidence, the problem
of how we cognize our own selves continues to lie faBow. 3
When living in this mode of experience, we perceive the present as a point which
divides two voids of non-being - the past and the future. In this sharp, punctual
section would have to constitute itself not only the actual 'I' with its whole life, but
also time itself. But - asks Ingarden - if I as areal subject arn supposed to be only
a sort of fictitious creation of my conscious processes, why should I not treat both
the past and the future as weIl as the strearn of time as a creation of my experience?
Such attempts have indeed been made in the history of philosophy, e.g. by Kant,
Bergson and Husserl. But it is difficult to understand how a constitution of anything
in a punctually understood present may be possible at all. Thus, this whole idea of
time becomes doubtful.
In order not to abandon it altogether, we could - says Ingarden - agree that our
conscious acts transcend 'temporal points'. But this is tantamount to granting that
something can be transcendent in this way and extend in time; then why not to agree
that all the beings given in the first sort of time-experience, in particular the real
human person, exist independently of our conscious acts?
What is more, if past and future do not exist, speaking about the present becomes
meaningless. And if past and future are but fictions of our conscious life, there exists
no reason to regard the stream of time as something 'absolute'.
Thus, the first, 'realist' experience of time seems to be more trustworthy. But, in
order to finally accept it we need - writes Ingarden - analyse in greater detail that
what is given in experience as weIl as the progress of these experiences.
In the fourth section of the essay 'Man and Time', published in 1946 together with
sections V and vI", Ingarden discusses practical consequences of living in the
second, 'idealist' mode. He shows that, taken seriously, this way of approaching
time makes us lose our own identity, the identity which we can preserve if we are
able to remain faithful to our own nature by controlling ourselves and by building
ourselves up in a constant struggle with our destiny and also with ourselves. When
we manage to command time, it gives us a possibility 'to shape one'S own self as a
continually growing inner power.,5
At the very beginning of part V Ingarden simply writes: 'There are my free and
responsible deeds [... ] which spring forth from the deepest interior of the ego: 6
Thus, he starts here directly from a certain - and indubitably essential - feature of
the really existing human person, from the fact of human freedom, and he makes a
metaphysical statement in his own understanding of metaphysics as the science of
the essence of what exists.
The possibility of free acting presupposes the existence of a real human subject,
different from the 'stream of consciousness', from the pure ego of transcendental
phenomenology. As a man - says Ingarden - 'I am apower which wants to be free,
[ ... ]but [ ... ] which can last and be free only when it voluntarily gives itself over to
the creation of goodness, beauty and truth,.7 It is, then, above all our experience of
man as an acting person which - according to the Polish phenomenologist -
speaks against the idealist interpretation of time and of the existence in time of man
and of the world in which man lives.
140 ANDRZEJ POLTAWSKI
When acting, man experiences his own self as transcendent to his own
experience, as identical in the flow of time, developing in it and, thus, making itself
independent of it. But, if man is too weak, not mature enough to solve the practical
problems of life which demand certain moral resolutions and decisions, if he betrays
himself, he slowly disintegrates in time. It seems. then - writes Ingarden - that
there exist not only different ways of experiencing time, but also 'different times {... ]
as the different modes of the enduring of psychic subjects, or at least as different
correlates of man's ways of enduring and behaving'; although, on the other hand,
'the most obvious thing to do would be to accept only one world time for everyone
and everything,.8
The essential thought is here, according to Ingarden, 'that time is a derivative
phenomenon, dependent on the behavior of the human person and, more generally,
of what exists'. On the other hand, 'there would also be the influence of time, of its
passage, on what exists, and in particular on the inner cohesion and distinctness of
man, on his inner power, on what ultimately matters in reality'. Time would
'certainly not be [... ] a pure form of experience, but a certain real force within the
realm of what is actual', and also 'something which settles over what exists, and
which at the same time comprehends it in some peculiar way so that what exists is in
time. And it would be different depending on how that which exists is, how it
behaves, lives .. .'9
3. The link between responsibility and time exists also because of the relation of
human action with the causal order of the world. 11
In this way, the possibility and sense of responsibility implies a determinate
temporal structure of the world, a structure consistent with the realist view according
to which the acting subject endures in time in the real world.
In the first volume of Streit Ingarden also distinguishes two specific moments of
temporal existence, i.e. (1) jissuration which ensues from the constant passing to
a new actuality and (2) fragility of the existence of everything that is temporal, its
destructibility which assumes the form of mortality in the living creatures.
a now as a point boundary between past and future, as a position in the time
continuum' .14 The description of the other, 'idealist' mode of experiencing time is
ambiguous in a particular way. On the one hand, it seems to flow from the
destructive role of time in our everyday life in the world, from the temporal
development and deeline of man; on the other hand, this aspect of the human destiny
is described as if it took place - so to speak - in a pure flow of time, in the
Husserlian 'pure consciousness', 'absolute flow' or 'absolute subjectivity'.
Thus, aiming at a philosophical, phenomenological substantiation of
commonsense realism, Ingarden points out the aporiai to which leads the second,
'idealist' conception of time, the idea that only the actual moment of time really
exists. In this way, he proceeds parallel to Paul Ricoeur, who defined the task he had
set to hirnself in his comprehensive work about time in the following way:
We will have 10 show this impossibilily of pure phenomenology of time. By 'pure
phenomenology' I understand an intuitive apprehension of the structure of time, [an
apprehension] which could not only be isolated from the argumentations by means of
which phenomenology !ries 10 resolve the aporiai received from anterior tradition, but
which does not pay still more and more for its results by new aporiai [...] the endless
aporiai of pure phenomenology would be the price which has to be paid for any attempt
of making 10 appear time itself, the ambition which defines as pure the phenomenology
oftime'lS
Having embraced in substance the 'realist' conception of time as the time of life
and action of man in the real world, Ingarden also speaks about constitution in a
sense which conforms with this conception, namely as a development, a building
itself up by a living creature, in particular by man. We may find his description of
this constitution in the first volume of Der Streit (par. 30). He regards this
description as but a preliminary and provisional sketch, because a full discussion of
it would imply an analysis of the 'material' (qualitative) essence of the living
individual or aperson; it would belong, then, to 'material' ontology which has not
been fully elaborated by hirn.
A living creature as a whole in the basic sense of the word (Ingarden thinks that
there are no wholes in the right and proper sense of the word in inanimate matter) is
characterised by a particular system and an irreversible order of its changes (or, at
the very least, of a certain selection of those changes), an order which deterrnines its
growth and its growing old. This order makes such a creature much more integrated
than inanimate things. Moreover, there exists a very elose connection between, on
the one hand, the phases of life and of the processes of development which take
place in a living individual and, on the other, the enduring properties and changing
states of this individual. As a historical being, the living individual depends much
more on the passage of time than inanimate things.
But, contrary to superficial appearance, not everything can change in a living
individual. Precisely the existence of an ordered system of its changes implies a
nature which endures in time. Moreover, it is not only the general structure of
species and genera that is relevant here, but, above all, a certain typical feature of all
the modes of behavior important for the life of a living creature or of aperson.
Those modes of behavior and those qualities which appear in an individual during
its life must be grounded in a certain nueleus of the individual which, as such,
INGARDEN ON THE PROBLEM OF TIME 143
remains unchanged. Ingarden calls this nucIeus 'individual constitutive nature'. The
introduction of this concept is connected with the distinction of 'static' and
'dynamic' identity which describe the status of, respectively, absolutely constant
qualities of an individual and those properties of it which may be - existentially and
intuitively - present in a different way while their quality remains strictly the same.
It is just the dynamic identity which, according to the Polish phenomenologist, is
characteristic of the properties corresponding to the individual nature of a living
creature or aperson. This very nature is, as Ingarden thinks, a Gestalt quality.16
Together with the qualities which are in a necessary way determined by its
qualitative content, it does not constitute itself in time but is, on the contrary, the
basis of all possible constitution of the objects in question. A living creature has,
then, a stratified structure. What is constituted in it are but aI the manner and the
degree of the development of its constitutive nature, bl those of its properties which
depend on the degree of realisation and development of this nature, and cl those
among its inessential properties which - in their general type - must necessarily
belong to its full character.
Thus, the constitution of a living creature or a person is essentially different from
the constitution of a process because 1/ it concems solely a part of the individual,
not its whole, 21 it is based not only on phases of processes, but also on some
constant factors, Le., on the one hand, on the individual constitutive nature of the
being in question and, on the other, on objects enduring in time which belong to the
environment of it.
According to his realistic and dynamistic attitude, the basic sense of
'constitution' is for Ingarden not a 'noematic' creation of units of sense realized by
'immanent' factors, but a development, a growth of 0 certain being itself.
While reflecting, in the respective parts of Streit, on the development of living
creatures and human persons, Ingarden does not say much about the difference
between those two sorts of real beings, but is mainly interested in their common
features. Nevertheless, he has a cIear idea of the particular character of the
development of man in contradistinction to, e.g., that of an animal. This is cIearly
shown in the last part of 'Man and Time' where he speaks about the creation of
goodness, beauty and truth as essential to human development. He also stresses the
dependence of subjective time on the attitude of the experiencing person, speaking
even about the possible necessity of accepting 'not only differing experiences 0/
time, but also, if one may put it so, different variants of time itself, correlative to the
different modes of my living and conduct,17 In Ingarden's lectures on ethics we may
also find some interesting and important remarks on the intrinsic relation between,
on the one hand, the moral conduct and development of a person and, on the other,
the crystallization of his or her inner organization and the particular dignity with
which the person is endowed owing to this organization. 18
Thus we can say that, analysing the development of living creatures and, above
all, of man, Ingarden has introduced an ontic conception of 'constitution', changing
radically the character of its problems. 19 But we may ask: what is, in the last res ort,
his attitude towards the 'pure phenomenology of time', the impossibility of which
Ricoeur tried to show?
144 ANDRZEJ PÖLTAWSKI
Thus Ingarden not only does not deny the possibility of an experience of pure
consciousness as taken in its 'purity' (be it only 'purely abstractively'), but he even
hopes to be able to construct on this experience a proof of the existence of the
world!
Nevertheless, to accept such an experience - be it only as a single object and be
it, even, not indubitable - seems tantamount to accepting the possibility of an
intuitive apprehension of time itself; this is so because pure consciousness is, as it
appears, understood by Husserl as something essentially temporal, one could say: as
time itself (or rather as something lurking behind time, constituting it - as 'absolute
subjectivity' - but to postulate a purely intuitive apprehension of such a
subjectivity seems to be even more absurd than to postulate it for time); and we have
no reason to regard Ingarden's idea of pure consciousness as essentially different
from that of Husserl's. Besides, pure consciousness seems to be, even for Ingarden, a
146 ANDRZEJ P6t.TAWSKI
rather mysterious being; this may be seen from his long deliberations on the
Husserlian 'phenomenological reduction' which is supposed to be the way to it and
on 'constitution'. Moreover, if our interpretation is accurate, it must be so because
'purity' of consciousness is, in fact, a phantom created by the desire to achieve - as
David Hume would say - certainty characteristic of 'relations of ideas' in the realm
of 'matters of fact' .
Ingarden's position concerning time is, then, ambiguous. On the one hand he
regards consciousness as an element of reality. This seems to imply the impossibility
of purely intuitive apprehension of consciousness and, consequently, the
impossibility of a pure apprehension of time. On the other hand he thinks that 'pure
consciousness' may be intuitively given; and this seems to be tantamount to
accepting an intuitive apprehension of the structure of time. This ambiguity is due to
an ambiguity of his general philosophical attitude, his attempt to connect realism
with the postulate of the existence of a particular sphere of absolute, final insights
i.e. with the application to consciousness of the Cartesian or Husserlian procedure of
deducing ontological or metaphysical assertions from epistemological postulates 27 .
And this ambiguity in its turn flows from the existence of two conflicting motives in
the philosophy of Edmund Husserl - the. critical and the dogmatic motive, stated
by Ernst Tugendhat; the first motive makes Husserl understand the purpose of
criticism of knowledge as a gradual explication and substantiation of it; the second,
dogmatic motive leads to the acceptance of the existence of a sphere of being the
investigation of which may lead to abs~lutely certain, final insights?8 As Tugendhat
shows, those two motives constantly interlace in Husserl's work. Ingarden took this
aporia from Husserl together with his theory of knowledge. This ambiguity is also
the root of the lack of clearness conceming the relation between ontology - as the
discipline treating of possible objects (or of the contents of ideas) - metaphysics
and experience in Ingarden's philosophy.29
Andrzej P61tawski
Bracka 1 m. 331-005, Cracow, Poland
E-mail: poltawski@hotmail.com
NOTES
1 'Der Mensch und die Zeit', Traveax du IX-e Congres International de Philosophie - Congres
Descartes. Paris 1937 vol. 1 p. 56-60.
2 The English text in: Roman Ingarden. Man and Value. 1983. Translation by A. Szylewicz. München
- Wien: Philosophia Verlag. p. 38.
3 Ibid .• p.42.
4 In Tw6rczosc. vol. 11 fase. 2 p. 121-137. reprinted in Ksiqieczka 0 czlowieku. 1964 Springfield [A ütlle
Book aboul Man]. 1972. Kfak6w.
S Man and Value. p. 48. italies of the original.
• Loc. eil.
7 Ibid .• p. 51.
8 Ibid.• p. 49. italies of the original.
9 Ibid.• p. 50. italics of the original.
10 First Polish edition: Roman Ingarden. Sp6r 0 istnienie swiata. 1947-48. Krak6w. 2nd Edition 1960/61.
Warszawa, German edition: Der Streit über die Existenz der Welt. 1965-66. Tübingen: Niemeyer Verl.
INGARDEN ON THE PROBLEM OF TIME 147
3'" Polish ed., taking account of the differences between the Polish and the Gerrnan text, 1987,
Warszawa. An English selection from the I" Vol: (I" Edition): Time and Modes of Being, trans!. by
Helen Michejda, 1964, Springfield, Ill. (Introduction, Ch. III, VI and Section 31 from Ch. VII).
11 Roman Ingarden, Über die Verantwortung. 1970, Stuttgart: Rec1am. English in Man and Value, op. cit.
12 Ingarden discusses the problem of constitution in Husserl in his lecture at the m International
Colloquium of Phenomenology in Royaumont in 1957 (Cahiers de Royaumont. Philosophie J/l.
Husserl, Paris 1959 p. 242-264).
13 Der Streit. op. cit., p. 206.
" 'On Responsibility. its Ontic Foundations' in Man and Value, p.l16, footnote 41.
15 Paul Ricoeur, Temps et reeit, 1983, Paris: ed. Seuil, vo!.l., p. 125: 'C'est cette impossibilite d'une
phenomenologie du temps qu'il faudra demontrer. Par phenomenologie pure, j'entands une
apprehension intuitive de la structure du temps, qui, non seulement puisse etre isolee des procedures
d'argumentation par lesquelles la phenomenologie s'emploie 11 resoudre les apories re~ues d'une
tradition anterieure, mais ne paie pas ses decouvertes par de nouvelles apories d'un prix toujours plus
eleve [... lles apories sans fin de la phenomenologie pure du du temps seraient le prix 11 payer pour toute
tentative de faire apparaftre le temps lui-meme, l'ambition qui definit comme pure la phenomenologie
du temps.'
16lntroducing 'Gestalt quality' as an ontological concept seerns to open Ingarden's ontology to the charge
of a certain sensualism. See my essays 'Wartosci a ontologia Ingardena [Values and Ingarden' s
Ontologyl ' in Roman Ingarden ajilozojia naszego ezasu [Roman Ingarden and the Philosophy of our
TimeI, A. W\;grzecki (ed.), 1995, Krak6w, Polskie Towarzystwo Filozoficzne, p. 111-122, and
'Problematyka doswiadczenia 'zewn\;trznego' w filozofii Romana Ingardena [The Problem of
'External' Experience in the Philosophy of Roman Ingarden)" 1996, Kwartalnik Filozojiezny (Krak6w)
XXIV: fasc. 3 p. 9-31.and fasc. 4, p. 97-123.
11 Man and Value, op. cit., p. 49.
18 Roman Ingarden, Wyklady z etyki [Lectures on Ethicsl 1989, Warszawa, p. 238 ff., 142f. See also my
'The Epistemological Locus of Moral Values' in Moral Truth and Moral Tradition, Luke Gorrnally
(ed.), Dublin and Portland, OR, Four Courts Press Ltd. p. 53-67, and also my 'Phenomenology and the
Status of Morality and Freedom' in Freedom in Contemporary Culture. Aets of the V World Congress
of Christian Philosophy. Catholie University of Lublin 20-25 Augst 1996. Vo!. 2. Lublin 1999,
University .Press of the Catholic University of Lublin p. 25-33.
19 See my 'Consciousness and Action in Ingarden's Thought' 1974 in Analeeta Husserliana. Vo!. III, p.
124-137.
20 Roman Ingarden, Der Streit, op. cit. Vo!. 11.2 p. 370-371: 'Das reine Bewusstsein scheint in concreto
im innersten Kern des realen Ichs enthalten zu sein und lässt sich nur rein abstraktiv, gewissermassen
rein gedenklich und nur bis zu einem gewissen Grade für sich abgrenzen. Ist es so, dann tritt es selbst
innerhalb der Welt als deren eigentümliches Element auf und ist aus dem ganzen Netz der weltlichen
Kausalzusammenhänge nicht herauszulösen, während es anderseits eben als "reines" konstituierendes
Bewusstsein außerhalb der Welt verbleiben und insbesondere von allem Zusammenhang mit dem
weltlichen Kausalnetz losgelöst sein müsste [... 1 Diese Loslösung und Abgrenzung ist nicht Sache der
transzendentalen Methode und überhaupt nicht bloß der Methode, die mehr oder weniger streng zu
befolgen ist. Es ist Sache des innigen Zusammenhanges zwischen dem "reinen" und dem "realen Ich",
welcher nicht erlaubt, bei der Verwendung der Methode streng zu verbleiben'. (italics of the original).
This text has been added in the Gerrnan edition, it does not exist in the first and second Polish edition.
21 Edmund Husserl, Ideas: Generallntroduetion to Phenomenology. translated by W.R. Boyce Gibson,
London: Allen&Unwin; New York: Macmillan p.153 (p. 93 ofthe Gerrnan original).
22 Loe. eit.
23 Loe. eil.
2. 'An object is existentially separate if. for ils existenee. it does not in its essenee require the existenee of
any other objeet with whieh it would have to eoexist. beeause of its essenee. within the eompass of one
and the same whole. In other words, if, owing to its essence, its existenee is not a neeessary eoexistenee
with some other objeet within a single whole . • (Time and Modes.. .. op. cit., p. 82).
25 •... it is possible that a certain object is existentially separate, and. in spite ofthis, in its essence requires
for its own existenee that of some other existentially separate object. We then say of it that it is
existentially eontingent upon it' (ibid., p. 89).
26 Streit. op. cit., o!. 11.2 p. 372: ' ... es eröfnet sich ein Gedankengnag, der - unter Mithilfe der material-
ontologischen Betrachtung - uns um einen wesentlichen Schritt der Lösung näherbringen kann. Ließe
148 ANDRZEJ PÖLTAWSKI
24. N. Rescher etal. (eds.), Essays in HonorolCarl G. Hempel. A Tribute on the Occasion ofHis
65th Birthday. 1969 ISBN 90-277-0085-0
25. P. V. Tavanec (ed.), Problems olthe Logic 01 Scientific Knowledge. Translated from Russian.
1970 ISBN 90-277-0087-7
26. M. Swain (ed.), Induction, Acceptance, and Rational Belief 1970 ISBN 90-277-0086-9
27. R. S. Cohen and R. J. Seeger (eds.), Ernst Mach: Physicist and Philosopher. [Boston Studies
in the Philosophy of Science, Vol. VI]. 1970 ISBN 90-277-0016-8
28. J. Hintikka and P. Suppes, Information and Inference. 1970 ISBN 90-277-0155-5
29. K. Lambert, Philosophical Problems in Logic. Some Recent Developments. 1970
ISBN 90-277-0079-6
30. R. A. Eberle, Nominalistic Systems. 1970 ISBN 90-277-0161-X
31. P. Weingartner and G. Zecha (eds.),/nduction, Physics, and Ethics. 1970 ISBN 90-277-0158-X
32. E. W. Beth, Aspects olModem Logic. Translated from Dutch. 1970 ISBN 90-277-0173-3
33. R. Hilpinen (ed.), Deontic Logic. Introductory and Systematic Readings. 1971
See also No. 152. ISBN Pb (1981 rev.) 90-277-1302-2
34. J.-L. Krivine, Introduction to Axiomatic Set Theory. Translated from French. 1971
ISBN 90-277-0169-5; Pb 90-277-0411-2
35. J. D. Sneed, The Logical Structure 01 Mathematical Physics. 2nd rev. ed., 1979
ISBN 90-277-1056-2; Pb 90-277-1059-7
36. C. R. Kordig, The lustification 01 Scientific Change. 1971
ISBN 90-277-0181-4; Pb 90-277-0475-9
37. M. Capek, Bergson and Modem Physics. AReinterpretation and Re-evaluation. [Boston
Studies in the Philosophy of Science, Vol. VII] 1971 ISBN 90-277-0186-5
38. N. R. Hanson, What I Do Not Believe, and Other Essays. Ed. by S. Toulmin and H. Woolf.
1971 ISBN 90-277-0191-1
39. R. C. Buck and R. S. Cohen (eds.), PSA 1970. Proceedings of the Second Biennial Meeting
of the Philosophy of Science Association, Boston, Fall 1970. In Memory of Rudolf Carnap.
[Boston Studies in the Philosophy of Science, Vol. VIII] 1971
ISBN 90-277-0187-3; Pb 90-277-0309-4
40. D. Davidson and G. Harman (eds.), Semantics 01 Natural Language. 1972
ISBN 90-277-0304-3; Pb 90-277-0310-8
41. Y. Bar-Hillel (ed.), Pragmatics 01 Natural Languages. 1971
ISBN 90-277-0194-6; Pb 90-277-0599-2
42. S. Stenlund, Combinators, 'Y Terms and ProolTheory. 1972 ISBN 90-277-0305-1
43. M. Strauss, Modem Physics and Its Philosophy. Selected Paper in the Logic, History, and
Philosophy of Science. 1972 ISBN 90-277-0230-6
44. M. Bunge, Method, Model and Matter. 1973 ISBN 90-277-0252-7
45. M. Bunge, Philosophy 01 Physics. 1973 ISBN 90-277-0253-5
46. A. A. Zinov'ev, Foundations olthe Logical Theory olScientific Knowledge (Complex Logic).
Revised and enlarged English edition with an appendix by G. A. Smirnov, E. A. Sidorenka, A.
M. Fedina and L. A. Bobrova. [Boston Studies in the Philosophy of Science, Vol. IX] 1973
ISBN 90-277-0193-8; Pb 90-277-0324-8
47. L. Tondi, Scientific Procedures. A Contribution concerning the Methodological Problems of
Scientific Concepts and Scientific Explanation. Translated from Czech by D. Short. Edited by
R.S. Cohen and M.W. Wartofsky. [Boston Studies in the Philosophy of Science, Vol. X] 1973
ISBN 90-277-0147-4; Pb 90-277-0323-X
48. N. R. Hanson, Constellations and Conjectures. 1973 ISBN 90-277-0192-X
SYNTHESE LIBRARY
90. J. Hintikka, The Intentions of Intentionality and Other New Models for Modalities. 1975
ISBN 90-277-0633-6; Pb 90-277-0634-4
91. W. Stegmüller, Collected Papers on Epistemology, Philosophy of Science and History of
Philosophy. 2 Volumes. 1977 Set ISBN 90-277-0767-7
92. D. M. Gabbay, Investigations in Modal and Tense Logics with Applications to Problems in
Philosophy and Linguistics. 1 9 7 6 · ISBN 90-277-0656-5
93. R. J. Bogdan, Locallnduction. 1976 ISBN 90-277-0649-2
94. S. Nowak, Understanding and Prediction. Essays in the Methodology of Social and Behavioral
Theories. 1976 ISBN 90-277-0558-5; Pb 90-277-1199-2
95. P. Mittelstaedt, Philosophical Problems ofModem Physics. [Boston Studies in the Philosophy
of Science, Vol. XVIII] 1976 ISBN 90-277-0285-3; Pb 90-277-0506-2
96. G. Holton and W. A. Blanpied (eds.), Science and Its Public: The Changing Relationship.
[Boston Studies in the Philosophy of Science, Vol. XXXIII] 1976
ISBN 90-277-0657-3; Pb 90-277-0658-1
97. M. Brand and D. Walton (eds.), Action Theory. 1976 ISBN 90-277-0671-9
98. P. Gochet, Outline of a Nominalist Theory of Propositions. An Essay in the Theory of Meaning
and in the Philosophy ofLogic. 1980 ISBN 90-277-1031-7
99. R. S. Cohen, P. K. Feyerabend, and M. W. Wartofsky (eds.), Essays in Memory ofImre Lakatos.
[Boston Studies in the Philosophy of Science, Vol. XXXIX] 1976
ISBN 90-277-0654-9; Pb 90-277-0655-7
100. R. S. Cohen and J. J. Stachel (eds.), Selected Papers of Uon Rosenfield. [Boston Studies in
the Philosophy of Science, Vol. XXI] 1979 ISBN 90-277-0651-4; Pb 90-277-0652-2
101. R. S. Cohen, C. A. Hooker, A. C. Michalos and J. W. van Evra (eds.), PSA 1974. Proceedings
ofthe 1974 Biennial Meeting ofthe Philosophy ofScience Association. [Boston Studies in the
Philosophy of Science, Vol. XXXII] 1976 ISBN 90-277-0647-6; Pb 90-277-0648-4
102. Y. Fried and J. Agassi, Paranoia. A Study in Diagnosis. [Boston Studies in the Philosophy of
Science, Vol. L] 1976 ISBN 90-277-0704-9; Pb 90-277-0705-7
103. M. Przel~ki, K. Szaniawski and R. W6jcicki (eds.), Formal Methods in the Methodology of
Empirical Sciences. 1976 ISBN 90-277-0698-0
104. J. M. Vickers, Belief and Probability. 1976 ISBN 90-277-0744-8
105. K. H. Wolff, Surrender and Catch. Experience and Inquiry Today. [Boston Studies in the
Philosophy of Science, Vol. LI] 1976 ISBN 90-277-0758-8; Pb 90-277-0765-0
106. K. Kosfk, Dialectics ofthe Concrete. A Study on Problems ofMan and World. [Boston Studies
in the Philosophy of Science, Vol. LII] 1976 ISBN 90-277-0761-8; Pb 90-277-0764-2
107. N. Goodman, The Structure of Appearance. 3rd ed. with an Introduction by G. Hellman.
[Boston Studies in the Philosophy of Science, Vol. LIII] 1977
ISBN 90-277-0773-1; Pb 9O-277-0774-X
108. K. Ajdukiewicz, The Scientific World-Perspective and Other Essays, 1931-1963. Translated
from Polish. Edited and with an Introduction by J. Giedymin. 1978 ISBN 90-277-0527-5
109. R. L. Causey, Unity ofScience. 1977 ISBN 90-277-0779-0
110. R. E. Grandy, Advanced Logicfor Applications. 1977 ISBN 90-277-0781-2
111. R. P. McArthur, Tense Logic. 1976 ISBN 90-277-0697-2
112. L. Lindahl, Position and Change. A Study in Law and Logic. Translated from Swedish by P.
Needham.1977 ISBN 90-277-0787-1
113. R. Thomela, Dispositions. 1978 ISBN 9O-277-081O-X
114. H. A. Simon, Models ofDiscovery and Other Topics in the Methods ofScience. [Boston Studies
in the Philosophy of Science, Vol. LN] 1977 ISBN 90-277-0812-6; Pb 90-277-0858-4
SYNTHESE LIBRARY
115. R. D. Rosenkrantz, In/erence, Method and Decision. Towards a Bayesian Philosophy of Sci-
ence.1977 ISBN 90-277-0817-7; Pb
90-277-0818-5
116. R. Thomela, Human Action and Its Explanation. A Study on the Philosophical Foundations of
Psychology. 1977 ISBN 90-277-0824-X
117. M. Lazerowitz, The Language 0/ Philosophy. Freud and Wittgenstein. [Boston Studies in the
Philosophy of Science, Vol. LV] 1977 ISBN 90-277-0826-6; Pb 90-277-0862-2
118. Not published 119. J. PeIe (ed.), Semiotics in Poland, 1894-1969. Translated from Polish.
1979 ISBN 90-277-0811-8
120. I. Pörn, Action Theory and Social Science. Some Formal Models. 1977 ISBN 90-277-0846-0
121. J. Margolis, Persons and Mind. The Prospects of Nonreductive Materialism. [Boston Studies
in the Philosophy of Science, Vol. LVII] 1977 ISBN 90-277-0854-1; Pb 90-277-0863-0
122. J. Hintikka, I. Niiniluoto, and E. Saarinen (eds.), Essays on Mathematical and Philosophical
Logic. 1979 ISBN 90-277-0879-7
123. T. A. F. Kuipers, Studies in Inductive Probability and Rational Expectation. 1978
ISBN 90-277-0882-7
124. E. Saarinen, R. Hilpinen, I. Niiniluoto and M. P. Hintikka (eds.), Essays in Honour 0/ Jaakko
Hintikka on the Occasion 0/ His 50th Birthday. 1979 ISBN 90-277-0916-5
125. G. Radnitzky and G. Andersson (eds.), Progress and Rationality in Science. [Boston Studies
in the Philosophy of Science, Vol. LVIII] 1978 ISBN 90-277-0921-1; Pb 90-277-0922-X
126. P. Mittelstaedt, Quantum Logic. 1978 ISBN 90-277-0925-4
127. K. A. Bowen, Model Theory /or Modal Logic. Kripke Models for Modal Predicate Calculi.
1979 ISBN 90-277-0929-7
128. H. A. Bursen, Dismantling the Memory Machine. A Philosophical Investigation of Machine
Theories ofMemory. 1978 ISBN 90-277-0933-5
129. M. W. Wartofsky, Models. Representation and the Scientific Understanding. [Boston Studies
in the Philosophy of Science, Vol. XLVIII] 1979 ISBN 90-277-0736-7; Pb 90-277-0947-5
130. D. Ihde, Technies and Praxis. A Philosophy ofTechnology. [Boston Studies in the Philosophy
of Science, Vol. XXIV] 1979 ISBN 90-277-0953-X; Pb 90-277-0954-8
131. J. J. Wiatr (ed.), Polish Essays in the Methodology o/the Social Sciences. [Boston Studies in
the Philosophy of Science, Vol. XXIX] 1979 ISBN 90-277-0723-5; Pb 90-277-0956-4
132. W. C. Salmon (ed.), Hans Reichenbach: Logical Empiricist. 1979 ISBN 90-277-0958-0
133. P. Bieri, R.-P. Horstmann and L. Kriiger (eds.), Transcendental Arguments in Science. Essays
in Epistemology. 1979 ISBN 90-277-0963-7; Pb 90-277-0964-5
134. M. Markovic and G. Petrovic (eds.), Praxis. Yugoslav Essays in the Philosophy and Methodo-
logy of the Social Sciences. [Boston Studies in the Philosophy of Science, Vol. XXXVI] 1979
ISBN 90-277-0727-8; Pb 90-277-0968-8
135. R. W 6jcicki, Topics in the Formal Methodology 0/ Empirical Sciences. Translated from Polish.
1979 ISBN 90-277-1004-X
136. G. Radnitzky and G. Andersson (eds.), The Structure and Development 0/ Science. [Boston
Studies in the Philosophy of Science, Vol. LIX] 1979
ISBN 90-277-0994-7; Pb 90-277-0995-5
137. J. C. Webb, Mechanism, Mentalism and Metamathematics. An Essay on Finitism. 1980
ISBN 90-277-1046-5
138. D. F. Gustafson and B. L. Tapscott (eds.), Body, Mind and Method. Essays in Honor of Virgil
C. Aldrich. 1979 ISBN 90-277-1013-9
139. L. Nowak, The Structure o/Idealization. Towards a Systematic Interpretation of the Marxian
Idea of Science. 1980 ISBN 90-277-1014-7
SYNTHESE LIBRARY
140. C. Perelman, The New Rhetorie and the Humanities. Essays on Rhetoric and Its Applications.
Translated from French and German. With an Introduction by H. Zyskind. 1979
ISBN 90-277-1018-X; Pb 90-277-1019-8
14l. W. Rabinowicz, Universalizability. A Study in Morals and Metaphysics. 1979
ISBN 90-277-1020-2
142. C. Perelman, lustiee, Law and Argument. Essays on Moral and Legal Reasoning. Translated
from French and German. With an Introduction by H.J. Berman. 1980
ISBN 90-277-1089-9; Pb 90-277-1090-2
143. S. Kanger and S. Öhman (eds.), Philosophy and Grammar. Papers on the Occasion of the
Quincentennial of Uppsala University. 1981 ISBN 90-277-1091-0
144. T. Pawlowski, Concept Formation in the Humanities and the Social Sciences. 1980
ISBN 90-277-1096-1
145. J. Hintikka, D. Gruender and E. Agazzi (eds.), Theory Change, Ancient Axiomaties and
Galileo's Methodology. Proceedings of the 1978 Pisa Conference on the History and Philosophy
of Science, Volume I. 1981 ISBN 90-277-1126-7
146. J. Hintikka, D. Gruender and E. Agazzi (eds.), Probabilistie Thinking, Thermodynamies,
and the Interaetion of the History and Philosophy of Scienee. Proceedings of the 1978 Pisa
Conference on the History and Philosophy of Science, Volume 11. 1981 ISBN 90-277-1127-5
147. U. Mönnich (ed.), Aspeets of Philosophieal Logie. Some Logical Forays into Central Notions
of Linguistics and Philosophy. 1981 ISBN 90-277-1201-8
148. D. M. Gabbay, Semantieallnvestigations in Heyting's lntuitionistie Logie. 1981
ISBN 90-277-1202-6
149. E. Agazzi (ed.), Modem Logie - A Survey. Historieal, Philosophieal, and Mathematical Aspects
ofModem Logic and Hs Applications. 1981 ISBN 90-277-1137-2
150. A. F. Parker-Rhodes, The Theory of Indistinguishables. A Search for Explanatory Principles
below the Level of Physics. 1981 ISBN 90-277-1214-X
151. J. C. Pitt, Pietures, Images, and Coneeptual Change. An Analysis ofWilfrid Sellars' Philosophy
of Science. 1981 ISBN 90-277-1276-X; Pb 90-277-1277-8
152. R. Hilpinen (ed.), New Studies in Deontie Logie. Norms, Actions, and the Foundations of
Ethics. 1981 ISBN 90-277-1278-6; Pb 90-277-1346-4
153. C. Dilworth, Scientifie Progress. A Study Concerning the Nature of the Relation between
Successive Scientific Theories. 3rd rev. ed., 1994 ISBN 0-7923-2487-0; Pb 0-7923-2488-9
154. D. Woodruff Srnith and R. McIntyre, Husserl and Intentionality. A Study of Mind, Meaning,
and Language. 1982 ISBN 90-277-1392-8; Pb 90-277-1730-3
155. R. J. Nelson, The Logie ofMind. 2nd. ed., 1989 ISBN 90-277-2819-4; Pb 90-277-2822-4
156. J. F. A. K. van Benthem, The Logie of Time. A Model-Theoretic Investigation into the Varieties
of Temporal Ontology, and Temporal Discourse. 1983; 2nd ed., 1991 ISBN 0-7923-1081-0
157. R. Swinbume (ed.), Space, Time and Causality. 1983 ISBN 90-277-1437-1
158. E. T. Jaynes, Papers on Probability, Statisties and Statistieal Physies. Ed. by R. D. Rozenkrantz.
1983 ISBN 90-277-1448-7; Pb (1989) 0-7923-0213-3
159. T. Chapman, Time: A Philosophieal Analysis. 1982 ISBN 90-277-1465-7
160. E. N. Zalta, Abstract Objeets. An Introduction to Axiomatic Metaphysics. 1983
ISBN 90-277-1474-6
16l. S. Harding and M. B. Hintikka (eds.), Diseovering Reality. Ferninist Perspectives on Epistem-
ology, Metaphysics, Methodology, and Philosophy of Science. 1983
ISBN 90-277-1496-7; Pb 90-277-1538-6
162. M. A. Stewart (ed.), Law, Morality and Rights. 1983 ISBN 90-277-1519-X
SYNTHESE LIBRARY
163. D. Mayr and G. Süssmann (eds.), Space, Time, and Mechanics. Basic Structures of a Physical
Theory. 1983 ISBN 90-277-1525-4
164. D. Gabbay and F. Guenthner (eds.), Handbook of Philosophical Logic. Vol. I: Elements of
Classical Logic. 1983 ISBN 90-277-1542-4
165. D. Gabbay and F. Guenthner (eds.), Handbook of Philosophical Logic. Vol. II: Extensions of
C1assica1 Logic. 1984 ISBN 90-277-1604-8
166. D. Gabbay and F. Guenthner (eds.), Handbook of Philosophical Logic. Vol. III: Alternative to
C1assical Logic. 1986 ISBN 90-277-1605-6
167. D. Gabbay and F. Guenthner (eds.), Handbook of Philosophical Logic. Vol. IV: Topics in the
Philosophy ofLanguage. 1989 ISBN 90-277-1606-4
168. A. J. I. Jones, Communication and Meaning. An Essay in Applied Modal Logic. 1983
ISBN 90-277-1543-2
169. M. Fitting, ProofMethodsfor Modal and Intuitionistic Logics. 1983 ISBN 90-277-1573-4
170. J. Margolis, Culture and Cultural Entities. Toward a New Unity of Science. 1984
ISBN 90-277-1574-2
171. R. Tuome1a, A Theory of Social Action. 1984 ISBN 90-277-1703-6
172. J. J. E. Gracia, E. Rabossi, E. Villanueva and M. Dascal (eds.), Philosophical Analysis in fAtin
America. 1984 ISBN 90-277-1749-4
173. P. Ziff, Epistemic Analysis. A Coherence Theory of Know1edge. 1984
ISBN 90-277-1751-7
174. P. Ziff, Antiaesthetics. An Appreciation of the Cow with the Subtile Nose. 1984
ISBN 90-277-1773-7
175. W. Balzer, D. A. Pearce, and H.-J. Schmidt (eds.), Reduction in Science. Structure, Examp1es,
Philosophical Problems. 1984 ISBN 90-277-1811-3
176. A. Peczenik, L. Lindahl and B. van Roermund (eds.), Theory of Legal Science. Proceedings of
the Conference on Legal Theory and Philosophy of Science (Lund, Sweden, December 1983).
1984 ISBN 90-277-1834-2
177. I. Niiniluoto, Is Science Progressive? 1984 ISBN 90-277-1835-0
178. B. K. Matilal and J. L. Shaw (eds.), Analytical Philosophy in Comparative Perspective. Explor-
atory Essays in Current Theories and Classical Indian Theories of Meaning and Reference.
1985 ISBN 90-277-1870-9
179. P. Kroes, Time: Its Structure and Role in Physical Theories. 1985 ISBN 90-277-1894-6
180. J. H. Fetzer, Sociobiology and Epistemology. 1985 ISBN 90-277-2005-3; Pb 90-277-2006-1
181. L. Haaparanta and J. Hintikka (eds.), Frege Synthesized. Essays on the Philosophical and
Foundational Work of Gottlob Frege. 1986 ISBN 90-277-2126-2
182. M. Detlefsen, Hilbert's Program. An Essay on Mathematical Instrumentalism. 1986
ISBN 90-277-2151-3
183. J. L. Golden and J. J. Pilotta (eds.), Practical Reasoning in Human Affairs. Studies in Honor
of Chaim Perelman. 1986 ISBN 90-277-2255-2
184. H. Zandvoort, Models ofScientific Development and the Case ofNuclear Magnetic Resonance.
1986 ISBN 90-277-2351-6
185. I. Niiniluoto, Truthlikeness. 1987 ISBN 90-277-2354-0
186. W. Balzer, C. U. Moulines and J. D. Sneed, An Architectonic for Science. The Structuralist
Program. 1987 ISBN 90-277-2403-2
187. D. Pearce, Roads to Commensurability. 1987 ISBN 90-277-2414-8
188. L. M. Vaina (ed.), Matters of Intelligence. Conceptual Structures in Cognitive Neuroscience.
1987 ISBN 90-277-2460-1
SYNTHESE LIBRARY
218. M. Kusch, Foucault's Strata anti Fields. An Investigation into Archaeological and Genealogical
Science Studies. 1991 ISBN 0-7923-1462-X
219. C.J. Posy, Kant's Philosophy ofMathematics. Modem Essays. 1992 ISBN 0-7923-1495-6
220. G. Van de Vijver, New Perspectives on Cybemetics. Self-Organization, Autonomy and Con-
nectionism.1992 ISBN 0-7923-1519-7
221. J.C. Nylri, Tradition anti Individuality. Essays. 1992 ISBN 0-7923-1566-9
222. R. Howell, Kant's Transcendental Deduction. An Analysis of Main Themes in His Critical
Philosophy.1992 ISBN 0-7923-1571-5
223. A. Garcfa de la Sienra, The Logical Foundations ofthe Marxian Theory ofValue. 1992
ISBN 0-7923-1778-5
224. D.S. Shwayder, Statement anti Referent. An Inquiry into the Foundations of Dur Conceptual
Order. 1992 ISBN 0-7923-1803-X
225. M. Rosen, Problems of the Hegelian Dialectic. Dialectic Reconstructed as a Logic of Human
Reality.1993 ISBN 0-7923-2047-6
226. P. Suppes, Models anti Methods in the Philosophy of Science: Selected Essays. 1993
ISBN 0-7923-2211-8
227. R. M. Dancy (ed.), Kant anti Critique: New Essays in Honor ofw' H. Werkmeister. 1993
ISBN 0-7923-2244-4
228. J. Wolenski (ed.), Philosophical Logic in Poland. 1993 ISBN 0-7923-2293-2
229. M. Oe Rijke (ed.), Diamonds anti Defaults. Studies in Pure and Applied Intensional Logic.
1993 ISBN 0-7923-2342-4
230. B.K. Matilal and A. Chakrabarti (eds.), Knowingfrom Words. Western and Indian Philosophical
Analysis of Understanding and Testimony. 1994 ISBN 0-7923-2345-9
231. S.A. Kleiner, The Logic ofDiscovery. A Theory of the Rationality of Scientific Research. 1993
ISBN 0-7923-2371-8
232. R. Festa, Optimwn Inductive Methods. A Study in Inductive Probability, Bayesian Statistics,
and Verisimilitude. 1993 ISBN 0-7923-2460-9
233. P. Humphreys (ed.), Patrick Suppes: Scientific Philosopher. Vol. 1: Probability and Probabilistic
Causality. 1994 ISBN 0-7923-2552-4
234. P. Humphreys (ed.), Patrick Suppes: Scientific Philosopher. Vol. 2: Philosophy of Physics,
Theory Structure, and Measurement Theory. 1994 ISBN 0-7923-2553-2
235. P. Humphreys (ed.), Patrick Suppes: Scientific Philosopher. Vol. 3: Language, Logic, and
Psychology. 1994 ISBN 0-7923-2862-0
Set ISBN (Vols 233-235) 0-7923-2554-0
236. D. Prawitz and D. Westerstähl (eds.), Logic anti Philosophy of Science in Uppsala. Papers
from the 9th International Congress of Logic, Methodology, and Philosophy of Science. 1994
ISBN 0-7923-2702-0
237. L. Haaparanta (ed.), Mind, Meaning anti Mathematies. Essays on the Philosophical Views of
Husserl and Frege. 1994 ISBN 0-7923-2703-9
238. J. Hintikka (ed.), Aspects of Metaphor. 1994 ISBN 0-7923-2786-1
239. B. McGuinness and G. Oliveri (eds.), The Philosophy ofMichael Dwnmett. With Replies from
Michael Dummett. 1994 ISBN 0-7923-2804-3
240. D. Jamieson (ed.), Language, Mind, anti Art. Essays in Appreciation and Analysis, In Honor
ofPaul Ziff. 1994 ISBN 0-7923-2810-8
241. G. Preyer, F. Siebelt and A. illfig (eds.), Language, Mind anti Epistemology. On Donald
Davidson's Philosophy. 1994 ISBN 0-7923-2811-6
242. P. Ehrlich (ed.), Real Nwnbers, Generalizations ofthe Reals, anti Theories ofContinua. 1994
ISBN 0-7923-2689-X
SYNTHESE LIBRARY
243. G. Debrock and M. Hulswit (eds.), Living Doubt. Essays conceming the epistemology of
Charles Sanders Peirce. 1994 ISBN 0-7923-2898-1
244. J. Srzednicki, To Know or Not to Know. Beyond Realism and Anti-Realism. 1994
ISBN 0-7923-2909-0
245. R. Egidi (ed.), Wittgenstein: Mind and Language. 1995 ISBN 0-7923-3171-0
246. A Hyslop, Other Minds. 1995 ISBN 0-7923-3245-8
247. L. P610s and M. Masuch (eds.), Applied Logic: How, What and Why. Logical Approaches to
Natural Language. 1995 ISBN 0-7923-3432-9
248. M. Krynicki, M. Mostowski and L.M. Szczerba (eds.), Quantifiers: Logics, Models and Com-
putation. Volume One: Surveys. 1995 ISBN 0-7923-3448-5
249. M. Krynicki, M. Mostowski and L.M. Szczerba (eds.), Quantifiers: Logics, Models and Com-
putation. Volume Two: Contributions. 1995 ISBN 0-7923-3449-3
Set ISBN (Vols 248 + 249) 0-7923-3450-7
250. R.A Watson, RepresentationalldeasJrom Plato to Patricia Churr:hland. 1995
ISBN 0-7923-3453-1
251. J. Hintikka (ed.), From Dedekind to Gödel. Essays on the Development ofthe Foundations of
Mathematics. 1995 ISBN 0-7923-3484-1
252. A Wisniewski, The Posing oJ Questions. Logical Foundations of Erotetic Inferences. 1995
ISBN 0-7923-3637-2
253. J. Peregrin, Doing Worlds with Woms. Formal Semantics without Formal Metaphysics. 1995
ISBN 0-7923-3742-5
254. I.A Kieseppä, Truthlikeness Jor Multidimensional, Quantitative Cognitive Problems. 1996
ISBN 0-7923-4005-1
255. P. Hugly and C. Sayward: lntensionality and Truth. An Essay on the Philosophy of AN. Prior.
1996 ISBN 0-7923-4119-8
256. L. Hankinson Nelson and J. Nelson (eds.): Feminism, Science, and the Philosophy oJ Science.
1997 ISBN 0-7923-4162-7
257. P.I. Bystrov and Y.N. Sadovsky (eds.): Philosophical Logic and Logical Philosophy. Essays in
Honour of Vladimir A. Smimov. 1996 ISBN 0-7923-4270-4
258. A.E. Andersson and N-E. Sahlin (eds.): The Complexity oJCreativity. 1996
ISBN 0-7923-4346-8
259. M.L. Dalla Chiara, K. Doets, D. Mundici and J. van Benthem (eds.): Logic and Scientific Meth-
ods. Volume One of the Tenth International Congress of Logic, Methodology and Philosophy
of Science, Florence, August 1995. 1997 ISBN 0-7923-4383-2
260. M.L. Dalla Chiara, K. Doets, D. Mundici and 1. van Benthem (eds.): Structures and Norms
in Science. Volume Two of the Tenth International Congress of Logic, Methodology and
Philosophy of Science, Florence, August 1995. 1997 ISBN 0-7923-4384-0
Set ISBN (Vols 259 + 260) 0-7923-4385-9
261. A Chakrabarti: Denying Existence. The Logic, Epistemology and Pragmatics of Negative
Existentials and Fictional Discourse. 1997 ISBN 0-7923-4388-3
262. A Biletzki: Talking Wolves. Thomas Hobbes on the Language of Politics and the Politics of
Language. 1997 ISBN 0-7923-4425-1
263. D. Nute (ed.): DeJeasible Deontic Logic. 1997 ISBN 0-7923-4630-0
264. U. Meixner: Axiomatic Formal Ontology. 1997 ISBN 0-7923-4747-X
265. I. Brinck: The lndexical '1'. The First Person in Thought and Language. 1997
ISBN 0-7923-4741-2
266. G. Hölmström-Hintikka and R. Tuomela (eds.): Contemporary Action Theory. Volume 1:
Individual Action. 1997 ISBN 0-7923-4753-6; Set: 0-7923-4754-4
SYNTHESE LIBRARY