A Collection of Polish Works On Philosophical Problems of Time and Spacetime

A COLLECTION OF POLISH WORKS ON PHILOSOPHICAL PROBLEMS
OF TIME AND SPACETIME

SYNTHESE LIBRARY
STUDIES IN EPISTEMOLOGY,
LOGIC, METHODOLOGY, AND PHILOSOPHY OF SCIENCE
Managing Editor:
JAAKKO HINTIKKA, Boston University, U.S.A.
Editors:
DIRK VAN DALEN, University of Utrecht, The Netherlands

DONALD DAVIDSON, University ofCalifomia, Berkeley, U.S.A.
THEO A.F. KUIPERS, University of Groningen, The Netherlands
PATRICK SUPPES, Stanford University, Califomia, U.S.A.
JAN WOLENSKI, Jagiellonian University, Krak6w, Poland
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
Springer-Science+Business Media, B.Y.

A C.I.P. Catalogue record for this book is available from the Library of Congress.
ISBN 978-90-481-6039-6 ISBN 978-94-017-0097-9 (eBook)

DOI 10.1007/978-94-017-0097-9
Printed on acid-free paper
All Rights Reserved

© 2002 Springer Science+Business Media Dordrecht
Originally published by Kluwer Academic Publishers in 2002.
Softcover reprint ofthe hardcover Ist edition 2002
No part of this work may be reproduced, stored in a retrieval system, or transmitted
in any form or by any means, electronic, mechanical, photocopying, microfilming, recording
or otherwise, without written permission from the Publisher, with the exception
of any material supplied specifically for the purpose of being entered
and executed on a computer system, for exclusive use by the purchaser of the work.
CONTENTS
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
This is a collection of some works of Polish philosophers and physicists on

philosophical problems of time and spacetime. Without restricting the thematic
scope of the papers, the issue conceming objectivity of time flow runs as a uniting
thread through most of them. Partly it is discussed directIy, and partly the authors
focus on themes which are of paramount importance for one's attitude to that
question.
In the first six papers the authors deal with their topics against the background of
contemporary physics, its theories, its difficulties and discussed conjectures. For the
paper of S. Snihur that background is provided by everyday worId-outlook, and the
author discusses the problem of existence and character of the future in the light of
basic principles of cIassical logic. The paper of A. P61tawski, about the views of the
outstanding polish philosopher Roman Ingarden, enriches the thematic scope of the
coIIection introducing into it some questions from philosophical anthropology and
ethics.
JERZY GOLOSZ
MOTION, SPACE, TIME*.
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).
H. Eilstein (ed.). A Collection 0/ Polish Wor/es on Philosophical Problems

0/ Time and Spacetime. 1-22.
© 2002 Kluwer Academic Publishers.
2 JERZY GOl..OSZ
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.
According to relationists the adequate theory of motion should contain in its

equations only relative particle quantities, such as relative particle distances, relative
particle velocities, relative particle accelerations, etc. or should refer to some
structures, e.g. the inertial or' affine structures, which are determined by distribution
of mass in the Uni verse.
The relational conception of motion (REL) gives the relationist a choice between
two alternative strategies. The first one is the classical strategy. Its first consistent
representative was Ch. Huygens. 2 The second one was considered by Newton in his
early work De Gravitatione (about 1668), but he rejected it as inadequate. It was
later undertaken by Berkeley (1752) and Mach (1883). According to this strategy,
inertial forces are produced by relative motion of bodies with respect to the fixed
stars. That strategy corresponds to the so-called Mach's principle, which says, that
inertial frames are deterrnined by distribution of mass in the Uni verse. It was only
when the general theory of relativity (hereafter GTR) came into being that the
advocates of the Mach's principle seemed to acquire a chance of realization of that
strategy. 1 will demonstrate in my paper whether or not their hope was well-founded.
The advocates of absolute conception of motion, like Newton, would of course
deny (REL), endorsing the following claim:
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.
Spacetime properties, i.e. spacetime symmetries acknowledged by a given

absolutist would decide, which of these quantities are used in his theory of motion.
Because the demand to construct a relational theory of motion also imposes some
spacetime symmetries on the spatiotemporal quantities represented in a given
theory of motion, the controversy between the relational and the absolute conception
of motion is related to another problem under consideration, namely, what spacetime
symmetries should have to be accepted in an adequate theory of motion.
2. THE PRERELATIVISTIC PHYSICS

The choice of spacetime symmetries accepted in Galileo's first modem theory
of motion was determined by a significant discovery made by its founder:
Shut yourself up with some friend in the main cabin below decks on some large ship,
and have with you there some flies, butterflies, and other small flying animals. Have
a large bowl of water with some fish in it; hang up a bottle that empties drop by drop
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
where R aß is a constant orthogonal matrix, va = const and a, ß =1,2,3. Notation

in this formula (and other ones in this text) follows Einstein's summation
convention: if an index is repeated once at the lower level and once at the upper
level, the summation must be carried out over the wh oIe range of that index.
Newton's first law of dynamics, as we understand it now, says, that there exists
a preferred class of motions, called free motions 3 , and there exist preferred reference
frames, called inertial frarnes, relative to which the free motions are rectilinear and
uniform. Newton's laws of dynamics have the same form in each of these inertial
frames. Every inertial frame is related to any other by some (GAL) transformation,
passively interpreted as a coordinate transformation. The passively interpreted
transformation should be understood as a change from old to new coordinates while
the actively interpreted transformation means acting on a system of particles
to produce, for example, a rotation or translation, or a velocity boost of the system.
The equation of motion, covariant with respect to the Galilean group (GAL),
is expressed by Newton's second law:
(1)
(where m - mass of a particle, Pz - an impressed force, x a - location of the

particle).
The equation (1) says, that the acceleration d 2x a Idt2 of a particle is directly
proportional to the impressed force and inversely proportional to the mass of the
particle.
In Newtonian physics there is no possibility to link up the inertial structure with
the mass distribution in the Uni verse, so we must attribute it to space and time.
4 JERZY GOLOSZ
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:
where M -' differential manifold, Ai - absolute objects and Pi - dynamical

objects.
We will define now the group of spacetime symmetries of a theory as the group
of all automorphisms of the absolute objects Ai of the theory i.e. the group of all
diffeomorphisms ':P that map M onto M in such a way that ':P*A i = Ai for all i. 6
The group of spacetime symmetries of Newtonian mechanics is the Galilean
group (GAL). We have the following absolute objects in this theory: flat affine
connection r ijk , time metric ti (representing absolute time) and Euclidean space
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)
The introduction in Newton's Scholium, of a distinguished reference frame into

the absolute spacetime structure means the necessity to restrict its symmetries by
6 JERZY GOLOSZ
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)
t ~ t' =/(t) , dfldt >0
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.
Leibnizian symmetries, in turn, have the form:
(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)
where F= (4<jmjmj rj/ 2/12, 'P= 4<jmj mjlrjj,

mj - mass of the i-th particle (1: mj = M), riJ{A) - relative particle distance,
rjj' = d rijl dA - relative particle velocity and A - an arbitrary time parameter.
The Lagrange function has a product form to ensure the independence of LdA
of the time parameter. There are only relative particle distances and relative particle
velocities in the Lagrange function L.
The, equations of motion are obtained by Barbour in the case of one dimension
(the case of three-dimensional space is omitted in the paper (1974) ) with the help
of Euler-Lagrange equations. If only a few particles are present in the Uni verse,
these equations lead to a motion which is very different from that prescribed by
Newtonian theory. But things look better when the environment is very similar
to ours - when there are very many particles (stars) distributed uniforrnly over
a large region. 'P is then effectively constant, time parameter A is indistinguishable
from Newtonian time and equations of motion take the form:
mjd x;' I dt = ( 11 M'P) . fJ'Pl fJXj (4)
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).
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.
3. THE RELATIVISTIC PHYSICS

In STR, Einstein assumed that light had the same velocity in each inertial frame and
also proposed a new principle of relativity. This principle, called now the special
10 JERZY GoLosz
principle of relativity, tells that the laws of physics (the equations

of electrodynarnics included) have the same form in each inertial frame. Basing on
these assumptions, Einstein proved that the absolute simultaneity should be replaced
by a relative simultaneity (with respect to a given reference frame) and introduced
formulas which correlate space and time coordinates in different reference frames.
These formulas tumed out to be identical with those of Larmor and Poincare, which
means that Maxwell's electrodynarnics satisfies the special principle of relativity.
In their most general form the transformations, discovered by Larmor, Poincare
=
and Einstein, are called Poincare transformations 11 and have the form (i, j, k, I 1, 2,
3,4):
(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)
where Xl =ct, ~ =X, ~ =y, x4 =z.

The spacetime interval LI .. resembles the distance in ordinary three-dimensional
Euclidean geometry (Llr2 =( X - Xo ! + ( y - Yo! + ( Z - Zo j), however, unlike the
Pythagorean square of distance. the square of the interval LI .. can also be negative.
Relative to a given point 0 (to. Xo. Yo. Zo) the values of the square of the interval
LI .. divide Minkowski spacetime into three disjoint classes of spacetime points.
The first one consist of points with lightlike (or null) separation from the point
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
LiT = .Jc LJ!,2 =cL1t'

2 (6)
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-v2 fe 2dt' (9)

to
12 JERZY GOI:.OSZ
where v is the velocity of the particle with respect to S.

The parameter '!' measured along the worldline of the particle is the proper time
of this particle and it is the time that would be measured by the clock moving
together with the particle. If we compare two objects, one of them at rest in Sand
the other sent for a joumey (v ~ 0) and then retumed to meet the former, we obtain
using (9):
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)
where / is the energy-momentumJour-veetor of the particle, mo the rest mass of this

particle and '!' the proper time. Just as Newtonian laws, this equation is valid only
in inertial reference frames. In noninertial reference frames additional terms
expressing pseudo forces, e.g. centrifugal or Coriolis forces, must appear in the
above equation. The equation of motion takes then the generally covariant form
(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
above, according to the Noether's theorem, every spacetime symmetry corresponds

with a conservation law. So the invariance under spatiotemporal translation
corresponds with the energy-momentum conservation law, the invariance under the
spatial rotation entails the angular momentum conservation law and the invariance
of STR under the Lorentz group entails the center of rnass conservation law (the
center of mass of an isolated system is in rectilinear uniform motion).
Einstein did not like STR I6 for two reasons. Firstly, no theory of gravitation
could be induded in STR in a satisfactory way. Secondly - and it was Mach's
influence - STR introduces the counterpart of the Newtonian absolute space in the
form of the dass of the inertial frames; the inertial frames affect the motion of
bodies but are not affected in turn. One could try two alternative ways to eliminate
such an absolute space. One way would consist in treating the inertial structure
assumed in STR as a dynamical element dependent on mass distribution (though not
determined by it). The other, more ambitious way consisted in an attempt to satisfy
the Mach's principle which asserts that inertial frames are determined by the
distribution of matter in the Universe. Einstein chose the second way.
It is worth noting, that although the two postulates pointed out above may seem
similar, they are based on different philosophical assumptions concerning spacetime.
The former teIls that distribution of matter in the Uni verse only affects the inertial
structure of spacetime, but not determines it. Because spacetime too, through its
inertial structure affects bodies (e.g. in the phenomenon of motion) we have a theory
which attributes an equal ontological status to spacetime and to the material world.
So the first postulate assumes that spacetime is a substance and leads to an absolute
theory of motion although it changes the absolute character of the inertial (or affine)
structure into a dynamical one. The other postulate is philosophically very unlike
the former. Determination of inertial (or affine) structure by the global rnass
distribution would enable one not only to treat the inertial structure as a dynarnical
object but it would also represent a relational solution of the controversy concerning
the ontological status of spacetime as weIl as of the controversy between relational
and absolute conception of motion. Considering a generally antiabsolutist attitude
of Einstein, no wonder that he chose the latter postulate.
Einstein noticed that the equality of the gravitational and inertial mass implied
that we cannot distinguish by means of local experiments between local
gravitational forces appearing in an inertial reference frame and inertial forces
appearing in a reference frame moving with a uniform acceleration with respect
to an inertial frame. Such frames are physically equivalent. It implies that
the invariance of physical laws under (POINC) postulated in STR is to narrow and
invariance under nonlinear transformations of coordinates in the four-dimensional
continuum should be postulated. This way a new, general principle of relativity
came into being. This principle together with the assumption that in the
nonrelativistic limit the sought for field equations should turn into the equations
of Newtonian theory of gravitation, led Einstein to the new gravitational field
equations:
(13)
14 JERZY GOLOSZ
where Rj - Ricci tensor, R - Ricci curvature scalar, G - gravitation al constant,

c - velocity of light, Tij - energy-momentum tensor. These equations are second-
-order nonlinear equations for components of the metric tensor 8ij . Since this tensor
determines the spacetime geometry while the tensor Tij appearing on the right-hand
side represents energy and momentum of physical systems, the field equations (13)
determine the effect of the distribution and motion of matter on spacetime geometry.
Although spacetime is curved, locally it takes the form of Minkowski spacetime and
thus it is legitimate to say that it is locally flat.
It is seen that spacetime with the metric tensor wh ich deterrnines it ceases to be
an absolute element in GTR. There thus are no absolute objects in GTR and the
symmetry group of this theory is the group of all differentiable transformations (C).
The last sentence expresses precisely the content of the general principle
of relativity.
It follows from the above considerations that by means of change of the absolute
character of the metric into the dynamical one Einstein implemented the weaker
of the two antiabsolutist programs discussed above. He intended, however,
to implement the other, more ambitious one, which was expressed by Mach's
principle. Did he succeed?
After finding the field equations in 1915, Einstein thought that he was
successful. However, it turned out quite soon that Mach's principle is not satisfied in
GTR. The deciding argument was provided by Wilhelm de Sitter in 1917. He found
for Einstein's field equations solutions describing the 'empty' space (Ti j = 0).17
It should be emphasized that emptiness assumed in the de Sitter's solutions does not
mean the absence of any bodies in the Universe, but only the absence of bodies
which would affect the large scale structure of spacetime. It thus can be assumed
that small test bodies exist in De Sitter's world and it should be possible to study
their inertia. We cannot suppose that their inertia is produced by interactions with
other celestial bodies, because this world is essentially empty. Since Einstein's
equations adrnit such a model of the Uni verse, Mach's principle is not satisfied in
GTR.
As far as the problem of motion is concerned, there is an important difference
between GTR and the older theories. In those theories equations of motion of field
sources were postulated independently of field equations, while in GTR the former
ones are the consequences of the latter. 18 This property of the field equations, which
manifests itself in their nonlinearity, was considered by Einstein (1949) very
important. He thought that a future, more general theory will have to keep this
property. Since in the first approximation the solutions of the equations of motions
in GTR coincide with the Newtonian ones,19 it can be expected that the relativistic
equations of motion also should be absolute. The analysis of the relativistic equations
of motion conforms this expectation.
GTR explains physical properties of motion in terms of geometrical properties
of curves in the spacetime manifold. 2o The affine connection introduced on the
manifold divides all motions into two classes: inertial motions, whose trajectories
are geodesics of the affine connection, and noninertial (accelerated) motions, whose
trajectories are not geodesics. The geodesics, which are the trajectories of freely
falling particles, are deterrnined by the equations
(14)
This equation of motion becomes simplified in a local inertial (freely falling)

frame, in which r~k = O. It takes then the form
(15)
Since Mach's program remains unfulfilled, there is no possibility to connect the

local inertial frames and the affine structure of spacetime to the distribution of
matter in the Uni verse. Therefore we have to connect it with spacetime. It turns out
that GTR is an absolute theory of motion.
It follows from the above considerations that both in STR and in GTR motion is
absolute. This, of course, does not rule out the possibility to create some other, more
general theory, which would apply to motions of all possible velocities and have a
re1ational character. Such a possibility, however, is denied by Earman in both his
works conceming the phenomenon of motion (1989a, p.85, 1989b, p.l02). Basing
on Malament (1985), Earman presents the following argument against the possibility
of constructing a relational relativistic theory of motion. First of all, he assurnes that
every spacetime, which possesses a recognizably relativistic structure, has to have
a null cone structure, or has to divide all tangent vectors to the given spacetime
manifold M into three nonoverlapping and mutually exhaustive categories: timelike,
spacelike and null, or relation of causal connectibilityl must be definable for all its
points. These three stipulations are provably equivalent. As Malament notes, the null
cone structure fixes the spacetime metric up to a conformal equivalence 22 and the
existence of rotation is a conformal invariant (the rotation vector computed relative
to one metric in the conformal class is nonzero just in case it is nonzero relative to
every member of this class). From this premise, Earman draws a conclusion that 'in
any spacetime one would want to count as relativistic, there is an absolute notion of
rotation' (Earman 1989a, p. 85). It means, in his opinion, that there can be no
relational relativistic theory of motion:
[... ] it suffices to establish what classical absolutists desperately wanted to prove but
never could. namely. that the very idea of spacetime in its relativistic guise is
irreconcilable with a full-blown relational conception of motion. 23
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.
4. ONTOLOGICAL IMPLICATIONS OF THE CONTROVERSY BETWEEN

THE ABSOLUTE AND RELATIONAL CONCEPTIONS OF MOTION
In this section I would like to consider relations which connect the controversy
conceming the absolute or relational character of motion to the controversy
conceming the ontological status of spacetime. Let us assign the name
of substantivalism to the view that spacetime is a substance. This can be understood
in the following way:
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:
-REL=:> SUB (16)
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
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:
-SUB~ REL (17)
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
differentiable transformations. These symmetry transformations make it possible

to introduce the corresponding spacetime structures and to divide them into absolute
objects (invariants of the symmetry transformations) and dynamical ones.
The fundamental fact is that in the light of the theory of relativity we have to
abandon the view that space and time are independent; we have to consider them
to be united into one entity - Jour-dimensional spacetime.
In this paper I also argued in favor of the thesis that all above considered theories
of motion are absolute and this entails the substantival character of space (in
Newtonian theory) or spacetime (in Relativity). I also tried to demonstrate that
Earman's attempt to avoid this inference by means of the representational ploy,
inspired by Sklar's proposal to treat absolute acceleration as a primitive monadic
property of particles, is implausible in the submitted form.
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.
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
QUANTUM SPACETIME AND THE PROBLEM

OF TIME IN QUANTUM GRAVITY
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.
1. WHY TO QUANTIZE GRAVITY?

Undoubtedly different people perceive time in different ways depending on their
culture and personality. However as a physicist I deeply believe that the irreducible
core of all distinct forms of time which exist in natural sciences and the humanities,
prose and poetry, philosophy and art, is provided by the physical time. The notion
of physical time was deeply modified in the first half of the 20th century due to
Einstein's relativity theory and now, at the turn of the rnillennia, the notion faces
again a necessity of a radical modification. An analogous modification is expected
to meet the concept of physical space. In both cases the modifications will occur
in the framework of quantum gravity.
What is quantum gravity? The term refers both to the whole
of quantum-mechanical effects in gravitational interactions and a theory describing
these effects. Quantum gravity is a curious branch of physics since up to the end
of 20th century no such quantum effects have ever been observed nor does a theory
of them exist. In a sense it is an empty notion. Yet the search for a theory
of quantum gravity has been the main line of research in fundamental physics for
the last two decades of that century and there is rather little doubt that it will be the
leading factor for the development of physics in the forthcoming decades. And it
should be explained to philosophers (while for physicists it is quite obvious) that the
quest för a theory of unobserved phenomena does not conflict with physics being
an empirical science.
23
H. Eilstein (ed.), A Collection 0/ Polish Works on Philosophical Problems
o/Time and Spacetime, 23-46.
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
appear. More specifically: what is quantum spacetime, and in particular, what

is time in quantum gravity?
The idea that gravity has ultimately a quantum nature was first expressed
by Einstein in 1918 in a paper on gravitational waves. For the next 30 years there
appeared only few occasional works on the subject and it was only in 1950 that Paul
Dirac began systematic investigations of quantization of GR. During the next half
century after his pioneering works most of eminent physicists working in elementary
particle physics, QF'T or GR endeavoured to construct a theory of quantum gravity
applying various approaches. It would be safe to state that this enormous effort of
a considerable number of highest quality scholars has produced up to now a very
modest outcome. At first the researchers have encountered very hard mathematical
difficulties. All mathematical problems already known in QFT turned out much
harder in quantum gravity. Furthermore a lot of new problems have been found.
Whilst dealing with these difficulties it turned out that in most cases they arise from
underlying conceptual problems. Gradually it was realized that almost all physical
concepts, which are precise and reliable in ordinary quantum physics and GR, lose
their precise character and actually make no sense when carried over to the domain
of quantum gravity. This is the case of spacetime and time evolution of a physical
system, probability, wave function (quantum state vector) and quantum
measurement. There is no stable ground on which theory of quantum gravity may be
based. All the attempts employing the methods developed within QFT have failed.
The two most promising approaches, which are based on string theory and on the
so--called loop quantum gravity (developing ideas introduced in 1987 by Abhay
Ashtekar) still remain in their initial stages and are coping with a multitude
of difficulties. They are very far from making any physical predictions that might
in principle be experimentally tested. Using anthropomorphic terms one would say
that gravity (particularly Einstein's GR) effectively protects itself from being
quantized.
The best example of the difficulties is the problem of time (and closely related
to it the problem of space). What is time in modem physics? As mentioned,
the concept of physical time is provided by GR. In the next section I present it and
in section 3 the troubles that the concept generates when quantum physics is taken
into account are discussed. The rather elaborated treatment of the concept of time
in GR is necessary to elucidate the problems dealt with in the last three sections.
2. TIME AND SPACE IN GENERAL RELATIVITY.
In GR time is not a fundamental entity, as it was according to Newton in non-

relativistic physics ('absolute, true and mathematical time, of itself, and from
its own nature, flows equably without relation to anything external and by another
name is called duration ... ' - Principia). Time is a secondary concept defined on
a fundamental object - spacetime, and since GR is a geometrically formulated
theory of gravity (or better: a physically interpreted curved spacetime geometry) this
concept refers to gravity and is expressed in terms of purely geometric quantities.
It should be stressed at the outset that in contrast with Newtonian physics and SR,
general relativity does not introduce a unique spacetime model but an infinite set of
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
be highly distorted in comparison to the genuine cone. We focus our attention on

curves having positive square of length and these are termed 'timelike curves'. As in
SR, all timelike curves through a point lie inside the null cone with its vertex at this
point.
For a body of negligibly small size its history (being the set of all elementary
events occurring to the body) describes a curved line in spacetime called 'worldline'
while a body of finite size describes a four-dimensional tube in spacetime. For
simplicity we will talk about pointlike bodies and their worldlines. Timelike curves
are physically distinguished in any spacetime, since it is a universal feature
of matter - established on purely empirical grounds - that all massive matter,
i.e. particles with rest mass different from zero, travel in the spacetime along
timelike worldlines. (Physicists often say that particles 'move' or 'travel' on
worldlines just to mean that a partide's history forms a curve in spacetime. This
'motion' is dearly different from the ordinary motion in physical 3-space.) There
are also particles with rest mass equal to zero (photons, gluons, neutrinos) whose
worldlines are null curves. This fact explains the name 'light cone' - photons
emitted radially from a pointlike lamp as a flash of light describe a null cone
in spacetime. Experiment then states that only timelike and null curves can be
worldlines for material bodies; spacelike curves are excluded.
For a physicist time is what is measured by a good dock. A dock is a device
applied to some cyclic physical process and counting successive cycles. A clock
is good if one can reasonably assurne that the periods of successive cycles are the
same (establishing this equality is a subtle problem and I will not deal with it. It is
sufficient to say that the sameness of periods can be reliably verified). A clock has
a mass, and thus assuming that it is small one may assign to it a single timelike
worldline. Time measured by a good dock following an arbitrary worldline is
regarded as the physical time related to this worldline. Defined in this way time has
a fundamental feature: it has a direct geometrical meaning - the physical time
which elapsed between two given events on the timelike worldline of the measuring
clock is equal to the length of this worldline (between these points) divided by the
light velocity c. This fundamental property of physical time was discovered by
Einstein while he was formulating SR. The property is carried over to the curved
spacetimes of GR. Thus the definition of physical time in both SR and GR is purely
geometrical and is the same. The geometrical definition of time is equivalent to its
definition in terms of a good dock since each timelike curve may be a worldline
of a dock and time measured by a good clock is independent of the physical
mechanism of the clock: all docks moving together in the same way (i.e. following
the same worldline) measure equal time intervals. That the flow of time is
independent of the physical nature of clocks was weil known in prerelativistic
physics. Without this property the notion of time would be meaningless in physics.
Relativity theory (both special and general) has introduced an essential novelty: time
is not a universal entity but is inseparably connected with timelike curves and is
defined for a given curve - only for adefinite timelike curve does it make sense to
speak quantitatively about the flow of physical time, i.e. about the length of a time
interval between two points on it. There exists nothing like 'time interval as such
between two points (events) in spacetime'. If one considers two different timelike
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.
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)
preferred in comparison to the others. For example: in SR by choosing an inertial

reference frame one defines a particular (global) division of the flat spacetime into
time and space and this splitting is physically superior to the splittings made with the
aid of non-inertial frames, but within the class of inertial frames no splitting is
preferred to the others. In GR the freedom of slicing spacetime into spacelike
hypersurfaces is far larger and all the splittings are on equal footing. A given
coordinate system uniquely determines the slicing but not the opposite; for a given
slicing a time coordinate is a parameter numbering the slices and one can freely alter
the numbering by means of any monotonous function. E.g. one can introduce a new
time coordinate as the logarithm of the previous one. Time coordinate is thus not
a genuine physical (measurable) quantity. Indeed, the fact that the time coordinate
can be freely changed by a time transformation (i.e. by reparametrizing the slicing)
without changing the studied geometrical (and physical) situation shows that the
concept of time coordinate has no deeper meaning. There is no other concept of time
which may be regarded as physical except that defined as the length of intervals of
a given timelike curve.
For the sake of completeness it is worth to comment on the idea, which is
favorite not only with philosophers - that of 'flow of time'. It has a precise
meaning in physics: each physical body (with a non-vanishing rest mass) has
a timelike worldline, and along that worldline both the physical time (calIed also
'the proper time' as it is measured by a clock staying at rest with respect to the body
and close to it) and the time coordinate in any coordinate system increase
monotonically. In other terms a body's worldline can be parametrized either by its
length (the physical time) or by an arbitrary time coordinate. Massless elementary
particles (photons etc.) describe null worldlines which by definition are of zero
length. Physically it means that there is no proper time for them (the photon cannot
be 'accompanied' by a clock). These worldlines may be parametrized by any time
coordinate. This fact provides a precise geometrical formulation of the intuitive
observation that 'one can stand still in space but one cannot stop moving in time'.
Time flow thus is universal, since for any physical object (massive or massless)
a time coordinate and the proper time (for massive ones) grow monotonically along
its worldline. Two aspects of this universality should be stressed. Firstly, it is a fact
that each worldline can be parametrized by a quantity possessing some features
which were traditionally assigned to the concept of universal time while the physical
time itself is not universal. Each object having a rest mass has its own time - its
proper time. On the other hand the coordinate time which is common for all
worldlines (in the piece of spacetime under consideration) is non-unique and is
unphysical (non-measurable). Secondly, the statement 'each kind of time flows for
every physical object' is based on a purely empirical fact that all worldlines are
either timelike or null and if one accepts this fact the statement geometrically sounds
almost trivial. The fact, however, that there are no spacelike worldlines in Nature is
not trivial. If there existed hypothetical elementary particles named 'tachyons',
whose squared mass were negative (their rest mass would be purely imaginary in the
sense of theory of complex numbers), they would move with respect to any inertial
reference frame (for simplicity we consider for the moment SR) at superluminal
velocities and their worldlines would be spacelike. Contrary to a common belief, the
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.
3. TIME AND QUANTUM THEORY. QUANTUM SPACETIME.

Although GR is a theory of the universal interaction - gravity - it remains apart
from the rest of physics. The concepts of space and time discussed in the previous
section are specific to GR and do not appear in other branches of physics. In aIl of
physics except GR time is viewed as a parameter serving as a measure
of change of any physical quantity. In non-relativistic physics time is an external
absolute parameter while in Special Relativity it is related to a given inertial
reference frame. However, though in SR time and space are transformed according
to Lorentz transformations while going from one inertial frame to another, in each
inertial frame time flows equably at each pi ace, since all clocks, which are at rest in
that frame and thus at rest with respect to each other, measure equal time intervals.
Since each inertial frame covers the entire spacetime (Minkowski space), in each
frame time is global and simultaneity of events is weIl defined. This allows to
conclude that modulo Lorentz transformations time is an external parameter for aIl
physical processes in the absence of gravitation.
Quantum theory as we know it (that is, both non-relativistic and relativistic
QFT) is based on this concept of time as a non-dynarnical quantity. This is
necessary for any probabilistic interpretation of quantum theory to work (not only
for the standard Copenhagen one but also for any other, e.g. for the so--<:alled
ensemble interpretation).
The probabilistic nature of the quantum world implies, in particular, that the state
or the physical situation of a quantum object - an electron, say - is described by a
wave function (which is currently termed 'state vector') which determines only the
probability of any possible outcome of any experiment (measurement) performed on

that object. Consider a simple experiment: one seeks for an electron in an otherwise
empty region of space which is accessible to it. The wave function then determines
the probability of finding it at a given place at a given time. The electron cannot
simultaneously be at two different places and simultaneous measurements may find
it at one place only. If the electron is sought for simultaneously in the whole
available space, it must be found somewhere; this means that the probability of
finding it somewhere is equal to one. This is a mathematical condition imposed on
the wave function; it is called 'the normalization of the state vector'. It is essential
that the experiment is performed everywhere at the same time.
The simultaneity of the measurements is crucial. And this simultaneity is with
respect to one physically distinguished time and not with respect to an arbitrary time
coordinate one might use in the spirit of GR. Otherwise one would arrive
at nonsensical, unphysical conclusions. In fact, assurne that the electron was
detected by a detector at point x of the spacetime. Being registered at x the electron
cannot be detected somewhere else later or at the same time, since it has been
captured and absorbed by the detector, and in practice it exists no more as aseparate
quantum particle. (To avoid confusion: in quantum measurement theory one
considers also 'non-demolishing experiments' in which the quantum object is not
destroyed and the measurement can be immediately repeated giving the same
outcome. A typical example is a measurement of spin of a quantum particle with the
aid of a magnetic field. Here we do not consider experiments belonging to this
class.) Thus the measurement process causes the reduction ('collapse') of the wave
function: it immediately (i.e. at infinite velocity) converges to x - for all spacetime
points later than and simultaneous with x the state vector vanishes (diminishes to
zero). If the state vector did not vanish immediately - if instead it was shrinking to
x for a finite time - then there would be a nonzero probability of finding the
electron at points that are later than x and spatially distant from it. Clearly this is
physically impossible since the electron is detected and disappears at x. (The fact
that the state vector converges to the spacetime point x with infinite velocity, or
equivalently, the fact that the information about the detection at x propagates
outwards immediately, is not in contradiction with the postulates of SR since the
wave function does not carry energy and its propagation does not correspond to
motion of a physical object. Thus it is not subject to relativistic restrictions.) The
concept of the reduction of the wave function as such is a non-relativistic concept,
though it may be reformulated to be compatible with Special Relativity. If instead of
this preferred concept of time one used another concept of time, the sets of
simultaneous events would be different and according to the other time the wave
function in general would be different from zero at points simultaneous to x!
Furthermore, in quantum theory for each measurable physical quantity
('observable') there exist specific state vectors, 'eigenvectors', such that this
observable has in these quantum states definite values: the eigenvector provides
certainty rather than a probability that in a measurement a specific value of the
observable will be found. The set of all observables referring to a given quantum
object consists of commuting and non-commuting quantities. The commuting
observables have a full system of common eigenvectors, while any two non-
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 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
not a theory of gravitational interactions taking place on a fixed spacetime

stage - here the stage itself is a dynamical quantity which actively participates in
all processes occurring on it. Gravity has no 'background' akin to that existing in all
other theories for all other physical processes. Gravity is its own stage and this is
regarded as a great Einstein's achievement. This is why the current attempts of some
field theorists to replace GR by string theory with strings existing in a fixed flat
multidimensional spacetime are considered by most relativists (and 'a relativist'
means here a physicist working in the realm of gravitational physics, not just an
adherent of Relativity Theory) as a step backwards with respect to Einstein's theory.
How to reconcile the absence of a fixed spacetime stage ('background') for
physical processes which is fundamental for GR with the necessity of having such
a rigid background in quantum theory? Is it possible to introduce a universal and
physical time, essential for the probabilistic interpretation of quantum theory,
if classical GR ascertains that no such entity exists? It is not known. What is worse,
the quantum fluctuations of the metric destroy, as I mentioned above, the absolute
partition of curves into timelike, spacelike and null ones and introduce instead
a probability of a given length, and consequently, a probability for any curve to
belong to one of these classes. Hence even the concept of the local geometrical
(proper) time cannot be fully maintained. (Here I assume the 'conservative'
approach to quantum gravity, in which the differential manifold structure remains
untouched and it is the metric that experiences fluctuations and is quantized. Curves
are then weIl defined and only their length is undetermined. As I already mentioned,
it is also possible to attempt to quantize the more fundamental structures of
spacetime and then the very concept of a curve may lose its meaning.) Therefore it
seems that among the first victims of quantization of gravity is time itself - it is no
longer a characteristic feature of some curves. The same fate meets the space.
The local splitting of spacetime into time and space is a classical notion and can be
carried over to a quantum theory of gravity only in a probabilistic way.
At first sight this change might seem quite logical and inevitable. Gravity is the
curvature of spacetime, therefore if one quantizes the former, time and space should
become quantum observables, just as, e.g., energy of an electron becomes converted
into a quantum quantity. The corresponding operators provide discrete (quantized)
values of time intervals and space volumes. However this operation is not harmless.
Space and time cannot be treated as merely two new quantum observables to be
added to the set of already known ones (energy, momentum, spin etc.). Due to
quantization any spacetime stage (fixed or dynamical) for quantum processes
disappears. Where do quantum gravity phenomena take place? And what does
quantum gravity have in common with quantum theory - as we currently
comprehend the latter?
In other terms: can we at least roughly say what is quantum spacetime? In my
opinion we cannot. In 1960's and 1970's it was widespread to envision quantum
spacetime using the concept of fluctuating topology, introduced by John A. Wheeler
in 1957. The basic idea, which neither he nor anyone else was capable to develop
into a theory, is following. Quantum fluctuations of the gravitational field, like those
of all other fields, increase at short distances (we assume that the concept of the
distance has a meaning) and at distances comparable to Planck length L p the
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
'foamlike' picture of spacetime is suggestive, it is now believed that it is too naive

and possibly not enough radical to be true. Either the conventional deseription of
spacetime as a smooth eontinuum is valid at all scales and topological fluetuations
simply do not exist or, more likely, quantum fluctuations of gravity are eompletely
different than in the 'foarnlike' picture and in order to deseribe them one needs
radically new ideas. At the end of 20th century the idea of quantum spacetime still
remains more a faney than adefinite mathematical and physical eoneept.
4. QUANTUM THEORY ON A FIXED CURVED SPACETIME

The above difficulties in quantizing the gravitational field (as weil as many other
reasons, which lie beyond the seope of this article) have suggested the view that we
should approach quantum gravity in small steps. Instead of constructing at onee
a full theory of quantum gravity one should successively solve simpler and less
ambitious problems. Suppose that we wish to find out a quantum meehanieal
description of some physical field, e.g. the electromagnetie one, in the presence
of a fixed gravitational field, i.e. assurning that the electromagnetie field does not
affect spacetime. Such physical situations are quite eommon (after all the entire
modern physics exeept GR neglects gravitational forces). Consider for instanee
a eurved spacetime of a black hole of Solar mass and a beam of light propagating in
it. The motion of light is strongly affected by the curvature while the gravitational
field of the light is negligibly weak and does not influenee the black hole. We look
for a fully quantum deseription of the light viewed as a swarrn of quanta of the
eleetromagnetic field. In other terms we wish to replace the ordinary QFf in
Minkowski spaee by an analogous QFf on an arbitrary fixed eurved spacetime - in
this case on the spacetime of a black hole. Attempts to construct such a theory were
initiated in early 1970's. After 1975 they were greatly intensified and stimulated by
the famous theoretieal discovery by Stephen Hawking of quantum particle creation
by black holes. He has found that though black holes are classical objects (in the
sense of being described by GR), they are intimately connected with quantum
effects: their classical gravitational field creates elementary particles in a purely
quantum-mechanical process. The Hawking's derivation of the 'black hole
radiation' was somewhat tricky and it became clear that the effect should be reliably
derived in the framework of QFf on the black hole spaeetime. In general, however,
the outeomes of the search for QFf on a curved spaeetime proved to be both
astonishing and disappointing. It turns out that most of the concepts and
developments of ordinary QFT cannot be carried over to a fixed curved spaeetime,
even for such a simple and physically significant case as that of a static spherically
symmetrie black hole (the Schwarzschild solution to Einstein's vaeuum field
equations). In spite of intensive investigations the theory still exists in an embryonie
form. The difficulties that have been encountered are best illustrated in the case of
two fundamental eoncepts of any quantum field theory: the vacuum state and the
quantum particle. In ordinary QFf the quantum vaeuum state is uniquely defined as
both the state of a quantum field having the lowest possible energy (not neeessarily
zero) and the state without real quanta (particles) of the field. The quantum particle
is mathematically defined as a certain solution of the field equations with positive
energy (while the corresponding solutions with negative energy represent

antiparticles). Physicists were amazed to find out that in the presence of a black hole
(and in other gravitational fields) there is no physically acceptable definition
of quantum vacuum. It turns out that for some models of spacetime the concept of
vacuum state depends on the choice of the observer (that is, of the reference frame):
different ob servers will perceive with their measuring devices different states as the
vacuum of the electromagnetic field, what is clearly physically inadrnissible. With
respect to some other models quantum vacuum just cannot be defined in a sensible
way. It is even worse with the concept of quantum particle: in the case of black
holes and other physically relevant spacetimes there are no solutions of the field
equations possessing properties fully corresponding to positive energy solutions in
flat spacetime. Therefore there is a number of physically inequivalent definitions
of particles.
The practical failure of search for QFf on a curved spacetime is regarded by
some authors as an indication that the aim is not attainable since a theory describing
all forms of matter as quantum objects while keeping gravity as a classical entity
would be inconsistent to some extent. These authors maintain that if matter is
quantized, then the gravitational field should be quantized too and a full theory of
quantum gravity is the only possible consistent theory for gravity and quantum
matter. May be they are right, but in my opinion the physical situation of a fixed
strong gravitational field (e.g. a black hole) and a portion of test quantum matter
(a beam of photons) moving in it is so unambiguous and clear that there should be
a unique theory describing it. The theory should arise from a quantum gravity theory
as a lirniting case for weak quantum gravitational effects. Perhaps while approaching
the theory from the opposite direction (having GR and the ordinary QFf as
a starting point) we make some mistakes. And it may be expected that the
difficulties of QFf in curved spacetimes reveal only a modest part of amazement we
will encounter in a full quantum gravity theory.
Surely the main lesson we have learnt from QFf in curved spacetimes is that
relativistic quantum theory - as we comprehend it - imposes very stringent
restrietions on the adrnissible spacetime. Nothing except the flat (Minkowski)
spacetime meets these requirements.
5. QUANTUM COSMOLOGY AND FOUNDATIONS

OF QUANTUM MECHANICS.
The physicist's views on the conceptual foundations of quantum theory (particularly

of non-relativistic quantum mechanics) have passed a characteristic evolution in the
second half of the 20th century. Initially it was commonly believed that all possible
doubtful and obscure issues in the subject were fully clarified in the famous debate
between Einstein and Bohr in 1920's and 1930's, and that the orthodox Copenhagen
interpretation of quantum mechanics is the ultimate interpretation of the theory.
Criticisms of the theory, expressed by Einstein and some others, were perceived as
due to conservatism and inability of the critics to comprehend and accept quantum
theory. At the same time GR was regarded as mathematically beautiful but
complicated and hard to grasp theory, and moreover as one which was of little
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
that it predicts probabilities of various effects, e.g. a probability that in a collision of

a photon with an electron a positron-electron pair will be created. In an ordinary
probabilistic theory its predictions are tested by registering relative frequencies of
various outcomes of experiments repeated on the corresponding statistical
ensembles, that is, on huge numbers of quantitatively identical situations, and by
comparing these statistical results with the probabilities predicted by the theory. Yet
the properties of the quantum vacuum state are not tested by performing
measurements on a statistical ensemble consisting of a great number of copies of
quantum vacuum. Properties of the vacuum are studied indirectly, via properties of
the quantum particles of the field.
The third objection to the Copenhagen interpretation is more essential. Due to its
nonlocality quantum mechanics is nonseparable. The fully free quantum particle,
which is perfectly isolated and separated from the rest of the Universe, is a mere
idealisation and, if an idealisation of that type might be innocuous in some contexts
in classical physics, it is not so in quantum mechanics. The wave function of each
particle extends over vast regions of space and thus wave functions of different
particles overlap, what implies some interactions between them. Clearly these
interactions are extremely weak and in the laboratory practice of experimental
physics they can be safely neglected, as they cannot be detected in a direct
experiment. Nevertheless they exist and they are responsible for 'quantum state
entanglement'. The dualistic partition of a studied situation into a quantum system
under consideration and its surrounding is only an approximation to the reality, for
no perfect boundaries separating the system from the surrounding can exist. This
means that if one studies, say, an electron, and assigns to it a one-particle wave
function, then it is only an approximate rather than exact description of the electron.
In fact, the real wave function depends on many other electrons and other particles.
In direct measurements the said approximation is fully satisfactory, yet as a matter
of principle one should not identify an approximate quantity with the accurate one.
Quantum state entanglement underlies some astonishing effects, e.g. the Einstein-
Podolsky-Rosen paradox, on which these authors based their criticism of
foundations of quantum mechanics in 1935.
Furtherrnore, due to the nonseparability of states in quantum mechanics the main
idea of Bohr, the founder of the Copenhagen interpretation, Le. the idea of
partitioning any physical situation into the quantum system (object) and the classical
observer (measuring device) is unsatisfactory. It is right and very useful as
a description of most laboratory experiments but as a fundamental description of the
Nature it is groundless. Nothing like a classical device exists: every instrument
consists of atoms andlor of elementary particles which are purely quantum objects.
A 'classical' object is actually a quantum one. A fundamental description of Nature
should be formulated solely in quantum terms and should be consistent and self-
contained. However, in spite of many intensive attempts, no such purely quantum
description of matter, making no reference to classical concepts, has been forrned.
Investigations of the last three decades of the 20'" century made no significant
progress in the subject and the Copenhagen interpretation of quantum mechanics
still dominates in practice. However many scholars are convinced that something
significant is missing in our understanding of the quantum world and that the
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
equations. However this overabundance of models makes some physicists feel

uneasy. In quantum cosmology the problem is far more severe. In ordinary quantum
mechanics a measurement made on all copies of a statistical ensemble is equivalent
to a measurement made on one physical object of a given type if it is repeated
(infinitely) many times. Thus instead of making a million identical measurements on
a million electrons in the same state, one may use only one electron and successively
make a million measurements on it, each time preparing the electron before the
experiment in the same quantum state. This is, however, possible providing there
exists an external observer capable of preparing the electron, with the aid of an
external device, many times in the same state. When the observer and the measuring
instrument are apart of the system (object) under investigation, any measurement
alters the state of the object and it is impossible to reproduce the original state 'from
inside'. Measurements on the Uni verse are unrepeatable. Can we therefore assign an
empirical meaning to the statement that the probability of this article having been
completed on October 30, 2000 equals 0,89? 3) In the study of the Uni verse there
disappears the distinction between interaction and measurement, while that
distinction is fundamental for the probabilistic interpretation of quantum theory.
In the case of an interaction within a quantum system as weIl as in the case of an
interaction of the system with other objects (e.g. the interaction of an electron with
an external magnetic field) the state vector of the system evolves deterministically
according to Schrödinger equation. (N.B. This means that in quantum theory its
probabilistic nature is completely contained in the state vector. Yet the time
evolution of the state vector is purely deterministic.) Yet the measurement
- performed by a classical device - causes an uncontrollable change of the state,
a change which cannot be predicted by quantum dynamics (i.e. by Schrödinger
equation). In the Copenhagen interpretation of quantum theory the measurement
process is the cause of the reduction of the wave function of the given quantum
object: the interaction with the classical instrument reduces in a random
(indeterministic) way the state vector of the object to one of the eigenvectors of the
measured quantity - the eigenvector corresponding to the value found in the
measurement. It is however unclear if and under what conditions the wave function
of the Uni verse is subject to the reduction to an appropriate eigenvector of the
measured observable. 4) Due to the absence of an external time the wave function of
the Uni verse does not describe it at one moment of time - i.e. it does not describe
the state of the space at a given time. Instead the wave function represents the state
of the entire spacetime - the past, the present and the future. This means that the
wave function does not evolve. The wave function of the Uni verse is a solution of
the Wheeler-DeWitt equation, a fundamental equation for quantum gravity
equation, where no time variable appears at aIl! The equation is regarded as an
'evolution' (or 'propagation') equation for the function, since the state vector is
determined by it, but actually in its view there is no time, no change, no evolution.
Besides these internal problems of quantum cosmology there is a problem of how to
connect the theory to the real Uni verse as we see it, Le. how to recover the ordinary
relativistic cosmology with the origin in the Big Bang and with the Universe passing
through the so-called inflationary evolution and finally forming the swarm of
receding galaxies.
Until the end of 201h century investigations of quantum cosmology, though

promising, made no significant progress. Quantum cosmology has not clarified the
foundations of quantum theory. It is likely that a breakthrough needs a radical
change of concepts while we are still imprisoned in a circle of traditional notions
and ways of thinking. In any case it is plausible to regard quantum gravity and
quantum cosmology as legitimate branches of physics.
In summary, the concepts of time, space and spacetime were deeply modified in
the first half of 201h century. In the second half of the century it turned out that these
concepts need a more radical modification in order to reconcile and unify classical
General Relativity with quantum theory. We still do not know what quantum
spacetime is, which traditional features of space and time will disappear and which
new amazing ones will arise when the unification will be achieved. It is safe to
presume that this will happen in a rather distant future and will significantly alter our
conception of the Nature.
Acknowledgement. This work was partially supported by a Polish Committee

for Scientific Research (KBN) grant no. 2P03D 01417.
Leszek M. Sokolowski
Astronomical Observatory,
Orla 171, 30-244 Cracow, Poland
E-mail: UFLSokol@TH.1F.UJ.edu.pl
MICHAL HELLER
TIME AND PHYSICS - A NONCOMMUTATIVE

REVOLUTION
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
Modern philosophy of time is inseparably connected with the progress in modern

physics. Newtonian mechanics and the idea of absolute time, classical thermo-
dynamics and the problem of the arrow of time, the study of non-linear dynamical
systems and the problem of the origin of irreversibility, theories of relativity (both
special and general) together with their revolutionary results concerning time
measurements - are but main headlines of this magnificent story on time and
physics. There are strong reasons to believe that the story is far from being
completed. The problem of time is intricately involved in currently discussed
interpretative issues of quantum mechanics (to mention ·only the measurement
problem: a sudden collapse of the unitary evolution to a fixed measurement result),
and it turns out to be one of the major stumbling blocks of almost all current
attempts to unify quantum physics with general relativity. The strategy adopted in
many recent models with respect to this problem is to eliminate it altogether by
claiming that there is no time on the fundamental level of physics. There are more
and more hints that this might indeed be the case. In the present study I shall not
deal with these fascinating issues; instead I shall try to introduce the reader into
a new field of mathematics, having broad spectrum of possible physical
applications, which in the near future can cause a new revolution in our
understanding of time.
If we look at the history of the problem, we can easily notice that the real
progress was always connected with beautiful mathematics. Newtonian mechanics
with its absolute time was based on beautiful mathematical structures (classical
dynamical systems, symplectic manifolds ... ), thermodynamics, both linear and non-
linear, was the result of beautiful mathematical structures (probability calculus,
stochastic processes ... ), the theories of special and general relativity are
implementation of beautiful mathematical structures (Minkowski and Lorentz
geometries), whereas all unresolved problems pertaining to time (e. g., the collapse
of the wave function in the act of measurement in quantum mechanics) arise on the
basis of some partial models and computational tricks. The latter ones are often
47
H. Eilstein (ed.), A Collection of Polish Works on Philosophical Problems
ofTime and Spacetime, 47-56.
48 MICHAL HELLER
met in various attempts to construct quantum gravity or final unifying theories.

In what folIows, I shall briefly present a relatively new mathematical theory, namely
noncommutative geometry, and discuss some of its applications to physics. It is
certainly a beautiful mathematical theory, and it is radical indeed. Being global from
the very beginning, it can be used to model an entirely new physics - physics with
no local concepts such as time instant or space point. It can mark another
breakthrough in our understanding of time and of the limitation of applicability
of this concept.
Noncommutative geometry has its sources in quantum physics where the family
of observables forms a noncommutative algebra, and in these branches of pure
mathematics in which analysis, algebra and geometry interact with each other. It has
emerged, however, as a more or less independent chapter of mathematics in the
works of Alain Connes (which are summarized in his book published in 1994).
The number of publications in this field, both in pure mathematics and in various
physical applications (e. g., gravity theories, unified gauge theories, classical
singularity problem, generalized Kaluza-Klein model), is rapidly growing (see the
books by Landi 1997 and Madore 1999). In the present paper, after briefly
presenting main ideas of noncommutative geometry, I shall explore its possibilities
to deal with these problems which could contribute to our understanding of time and
its origin.
The organization of the paper runs as folIows. In section 2 I show how the main
idea of noncommutative geometry emerges from previous investigations. The non-
local character of noncommutative spaces (briefly presented in section 3) allows one
(as it is demonstrated in section 4) to introduce a dynarnics without the usual
concept of time (a "timeless dynamics"). In section 5 I show that, by imposing some
additional constraints on a given noncommutative algebra, we can gradually
improve its temporal properties to finally obtain the usual time of classical physics.
It turns out (in section 6) that the standard dynamics, the emergence of time, and the
origin of probability have common roots in noncommutative geometry.
My interest in noncommutative geometry has also a philosophical motivation.
Progress in science is intrinsically linked with evolution of concepts, and many
concepts of great philosophical importance, such as the concepts of causality and
chance, undergo a radical change when passing from commutative geometry
to noncommutative geometry. This change consists in far-reaching generalizations.
It turns out that no standard concept of time and of event is necessary to make one
able to meaningfully speak of causality and chance. This is certainly an important
philosophical lesson: our seemingly most universal concepts have their limited
domain of applicability. Since this lesson is based on strict mathematical results and
not on any of their particular applications, it will remain valid even if it turns out
that these particular applications fail to model the fundarnentallevel of physics. This
question is discussed in section 7.
TIME AND PHYSICS 49
2. SOURCES OF NONCOMMUTATIVE GEOMETRY

Noncommutative geometry has two sources: quantum mechanics and problems that
arise at a rather fuzzy borderline between differential geometry, algebra and
functional analysis (see Manin 1991, pp. 3-8). It often happens in science that germs
of future achievements are present in well-known theories or models long before
somebody realizes that they are powerful enough to initiate a major breakthrough in
a certain field of research. For a long time it was very weil known that
noncommutativity plays an important role in quantum physics. If two observables do
not commute, the corresponding physical properties cannot be measured
simultaneously with any desired precision. For instance, this is valid as far as the
position and momentum observables are concerned. The famous Heisenberg
uncertainty relations are simple consequences of the noncommutativity of some
observables. Noncommutative geometry can be regarded as a reconstruction,
in geometrical terms, of a "Iogical scheme" underlying the aforementioned
properties. In this sense, the Planck constant can be regarded as a "deformation
parameter" from commutativity of cIassical physics to noncommutativity
of quantum theory.
Both in physics and in mathematics there is an ever-growing need for more and
more general spaces. In recent years physicists and mathematicians have often been
confronted with spaces which seem highly pathological; for instance, some spaces
studied by them are reduced, from the topological point of view, to a single point
(technically, they are called non-Hausdorff spaces). In such a case, one could follow
two strategies: either to regard such spaces as "meaningless" and banish them from
science, or to look for more powerful mathematical methods to cope with such
spaces. Noncommutative geometry is precisely the result of the latter approach. In
modern differential geometry the notion of a "weil behaved space" has assumed the
form of the concept of differential manifold (or manifold, for simplicity). Examples
of differential manifolds are: n-dimensional EucIidean space, Minkowski's
spacetime of special relativity and spacetimes of general relativity (if singularities
are not treated as parts of them) and many other spaces considered in physics. The
usual way of defining manifolds is in terms of coordinate systems, but it can be
shown that the entire information about a given differential manifold is contained in
the family of all smooth functions on this manifold. In fact, one can equivalently
define a manifold as the pair (M, C) where M is a non-empty set and C is the family
of aIl smooth functions on M. This family forms what mathematicians call
an algebra. This means that functions belonging to C can be added and multiplied
with each other and multiplied by scalars (numbers). Working with coordinates is
easier for practical (i.e., computational) purposes, but the method of the smooth
function algebras turns out to be better adapted for further generalizations. The main
idea is to replace a given functional algebra C by an (associative) algebra A. The
algebra Cis commutative because functions are multiplied in the commutative way,
but a general algebra need not be commutative, and this is how noncommutative
geometry is born.
50 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
time evolution of observables in quantum mechanics is also defined in terms of

unitaryoperators.
Now, the last step has to be made to obtain the correspondence with the ordinary
time of special or general theories of relativity. This step is rather simple. Among all
elements of the algebra A one can distinguish the sub set of those elements which
commute with all elements of A; this subset is called the center of A and is denoted
by Z(A). Of course, if we restrict the algebra A to its center (or to a non-empty subset
B of it), we obtain a commutative algebra, and the space (M, Z(A)) (or (M, B) is the
usual (commutative) space. Essentially, all concepts, that have been generalized
with the help of noncommutative geometry go over into their usual (i.e.,
commutative) 'special cases', if the algebra A is restricted to its center Z(A) or its
subset B.
6. TIME AND PROBABILITY

As we have seen, von Neumann algebras are 'dynarnical objects': they lead, in
a natural way, to the temporal ordering and to the dynamical evolution of considered
systems. They are also 'probabilistic objects'. Let us be more specific in this respect.
The concept of prob ability, in the usual sense, is a special instant of a more general
concept, namely that of measure, and the theory of probability is but a chapter of the
measure theory. Measure is a function on a farnily of subsets of aspace, called
measure space (the subsets are called measurable subsets); this function is assumed to
be real valued and positive. If the measure of the entire measure space is equal to 1,
the measure is called probability. This implies that the probability of any subset of the
measure space (called now the probability space) is contained between 0 and 1.
And the fact that the probability of the entire probability space is equal to 1 means that
'something must happen' (one says that probability is a measure normed to 1).
Let us notice that this concept of probability presupposes individual events - the
elements of the probability space. How could the probability concept be implanted into
the noncommutative environment in which there are no individuals? As we remember,
von Neumann algebra (roughly speaking) is a C·-algebra with a distinguished state
which, by definition, is positive and with unit norm. If we have in rnind that the state,
as a functional, generalizes the concept of function, we realize the dose affinity of
probability (which is a positive function normed to 1) with von Neumann algebras.
Indeed, von Neumann algebras can be thought of as generalizations of the standard
measure theory (see, for instance, Connes 1994, chapter I). As we can see, the concept
that generalizes the standard concept of probability can be applied to the mathematical
description of this level of physics on which the usual notion of time is meaningless.
We should notice a remarkable unifying power of noncommutative geometry: the
theory of von Neumann algebras unifies three dassically distinct things: time,
dynarnics and probability.
7. SOME CONSEQUENCES OF NONCOMMUTATIVE REVOLUTION

The evolution of concepts is both a driving force of scientific progress and one of its
main consequences. On the one hand, without the evolution of concepts science would
54 MICHAL HELLER
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
associated with probability and chance should be replaced by abstract algorithms.

Since von Neumann algebras are also 'dynamical entities', we can say that the
noncommutative dynamics and noncommutative probability are but two aspects of the
same thing. Chance can be regarded as an element of this game.
In the contemporary 'natural philosophy' there are strong tendencies to reduce all
explanations to a 'game of probabiJities'. According to a ~uite popular idea, on the
'fundamental level', or in the set of 'all possible universes' , 1 there reigns chance and
chaos, and the usual laws of nature emerge of them as the result of some averaging
processes (see, for instance, Smolin 1997). This kind of philosophy tacitly assurnes
that the usual concept of probability remains valid on all levels of physical reality.
But, as we have seen, there are strong reasons to believe that the usual concept
of probability is valid only within a certain domain, and beyond this domain it should
be replaced by its suitable generalizations.
We are indeed entitled to expect that the 'noncommutative revolution' will have
a powerful impact on many areas of philosophical speculations.
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
FLOW OF TIME AS A SELECTION RULE

IN GENERAL RELATIVITY
Abstract. Lorentz-covariance of special-relativistic fonnulae expresses adefinite spacetime symmetry,

which is sometimes claimed to rule out the possibility of time flow. Such an argument from spacetime
symmetry is strengthened by the general theory of relativity only if the general covariance of physical
fonnulae yields more spacetime symmetry. It is argued that no general-relativistic spacetime is more
symmetrie than the special-relativistic one. The opposite opinion results from confusing dynamicity of
spacetime metric with its relativity. Moreover, the whole argument from symmetry rests on confusing the
physical spacetime with the full spacetime structure of the world. On the other hand, the so-called hole
argument indicates some excess of generality in generally-covariant Einstein equations. The possibility of
time flow is proposed as aselection rule for eliminating non-physical solutions. It is argued that this
excludes no realistic cosmological models.
1. THE ARGUMENT FROM SPACETIME SYMMETRY
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
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.
2. RELA TIVITY IN SPECIAL AND GENERAL THEORIES

OF RELATIVITY
Ascribing to the spacetime of GTR a symmetry richer than the symmetry of the
spacetime of STR is seerningly supported by the fact that the spacetime pseudo-
metric (below, for brevity, albeit a bit misleadingly, called metric), which was
absolute in special-relativistic spacetime, is made by GTR adynamie element of
spacetime structure. On the other hand, aspacetime symmetry group can be defined
as a grou~ of transformations that preserve the absolute elements of the given
spacetime. However, it is relativeness that is usually considered an opposite to
absoluteness. Does it mean, thus, that every dynamic element of spacetime structure
is in some reasonable sense relative?
Relativity is usually understood as dependence on the point of view. Yet, such
description is ambiguous, as we will see later. In interpretations of relativity theory
it is commonly assumed that the point of view is determined by reference frame. For
it to be true, every physical quantity must have strictly the same operational meaning
for all observers. Then relativity becomes to meanframe-dependence.
Would it be dependence on any reference frame? As we know, some procedures
that define the meaning of such quantities as, e.g., time coordinate can give
inconsistent results when applied in non-inertial frames. 6 Ignoring this fact leads,
e.g., to the twin paradox and to a less known paradox connected with the so-called
Sagnac effect. 7 In the spacetime of STR the c1ass of inertial frames is, thus,
distinguished by the absence of effects that restrict applicability of those procedures.
Therefore, relativity in STR means dependence on choice not of an arbitrary, but of
an inertial frame of reference.
The general theory of relativity is a generalization of the special theory. Does its
higher generality mean introducing more re1ativity? Certainly higher relativity is not
its consequence, i.e. does not hold in all its models, since the spacetime of STR is
among them. But perhaps it, at least, enables more relativity? It would be so, if in
some general-relativistic cosmological models the scope of relative elements of
structure were broader than in the spacetime of STR.
However, what does it mean that some element of the structure of physical
spacetime is relative? As a good example, let us consider simultaneity in special-
relativistic spacetime. There is a procedure that defines in every inertial frame a two
FLoWOFTIME 59
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
3. GTR AND THE ARGUMENT FROM SYMMETRY

Let us remember that symmetries 0/ a theory are spacetime transformations that
preserve absolute elements of its structure. Such a transformation may be expressed
as a suitable coordinate transformation, interpreted actively.lO On the other hand, the
symmetries of spacetime of STR correspond to the principle of relativity that holds
in it. Consequently, each transformation which is its symmetry can be also
interpreted passively, as a change of the reference frame in which the spacetime is
described. Now, remember that relative character should be ascribed to the elements
of spacetime structure the description of which can change with a change of the
reference frame. Thus, it is natural to regard as spacetime symmetries the
transformations that preserve all its non-relative elements, rather than only the
absolute ones.
As we have already noticed, in the spacetimes of GTR the metric is non-relative.
The above resolution would mean, thus, that the symmetries of such aspacetime
must be its isometries. In the spacetime of STR, as a special case, these are all
transformations of the Poincare group. However, let us remember that it is an
'empty' spacetime, which does not contain any object endowed with a non-
negligible mass. In aspacetime that contains matter, a very special distribution of
the matter can make the spacetime to have some isometries, although one cannot
expect that the class of these isometries will be as rieh as the Poincare group.
In general, however, a GTR spacetime does not have to possess any non-trivial
isometries, and thus, according to our criterion, any symmetry.11
At first sight, this result seems to contradict the fact that the scope of elements of
spacetime structure that are regarded as absolute in the framework of GTR is
narrower than in STR, which would mean ascribing to the former theory a symmetry
richer than to the latter. However, the appearance of contradiction disappears if we
distinguish the symmetry of a theory from symmetries of its models. In the case of
STR they coincide, but there is no reason that it must always be so. To retain this
coincidence also in the framework of GTR, one would have to assume, as indeed
some authors dO,I2 that this theory ascribes to spacetime a poorer structure than the
structure of a pseudo-Riemannian space. However, the spacetimes of GTR are not
devoid of metric, although the metric is different in different spacetimes.
If the above considerations are correct, GTR not only does not impose upon
spacetime any symmetry richer than that of STR; on the contrary, it prohibits such a
richer symmetry. It means, of course, weakening, instead of the expected
strengthening, the argument against the flow of time from spacetime symmetry. And
let us remember that even with respect to the spacetime of STR, where there was the
most solid ground for it, its effectiveness was doubtful.
4. GTR AND RELATIVITY OF METRIC

Now, what should we do with the view that GTR introduces relativity
0/ geometry?13 In the light of the above analyses, such statement would be untenable
if relativity were to be understood just as dependence on reference frame. In order to
take it seriously, one would have to annul the above considered assumption that the
62 JAN CZERNIAWSKI
point of view the metric should depend on is unequivocally determined by the

reference frame. And as some ground for rejecting that assumption one might try to
make use of the interpretation of gravitation as a so-called universal force, 14 since
the unequivocality of the results of metric defining procedures may be undermined
by arguing that they depend on the influence of gravitation on physical standards of
length and time units.
Seen from this perspective, one should regard as 'true' the metric that may be
ascribed to spacetime on the basis of standard procedures as performed in the
absence of universal forces. The results of their application in gravitation al field
should, thus, have to be corrected in such way that the influence of gravitation on
the standards of units would be eliminated. However, the concrete corrections would
depend on the assumption about that influence. Consequently, one might ascribe to
spacetime different metrics from different points of view, which would be
determined not only by the reference frame, but also by the assumption concerning
the distribution of the universal force. 15 According to that approach, abstaining from
any corrections would correspond to the naive assumption that there is no universal
force and that thus the standard procedures define the 'true' metric of spacetime.
Yet, the following problem arises in connection with the above interpretation.
In the light of the equivalence principle there is no way of non-arbitrary deciding
between different assumptions about the distribution of the gravitational field as
an universal force. Nevertheless, some decision would be inescapable as to how the
results of measurement procedures shall be corrected. Among various possible
arbitrary decisions, the resolution not to correct them, no doubt, is clearly
distinguished from the pragmatic point of view.
Besides, such decision does not have to be motivated by any naive assumption,
but it may be understood as a conscious operation of geometrization of the universal
force. 16 Exactly in this way it is understood in the framework of the general theory
of relativity. Thus, the thesis of relativity of geometry proves to be a relic of non-
relativistic thinking, for on the basis of the interpretation of gravitation as universal
force, too, one must reconcile oneself with the fact that the 'true' metric is non-
physical, or, in other words, metaphysical, whereas the metric which is the result of
geometrizing gravitation has a well-defined operational meaning and is not relative.
Similar error may be committed in the framework of the interpretation of STR in
connection with the problem of operational meaning of simultaneity. The effects of
the 'ether wind' in reference frames moving relative to the ether may be treated as
the result of action of something like universal force, which would influence the
results of physical procedures defining simultaneity. Consequently, one may be
under the impression that in such frames the results of those procedures should be
corrected accordingly.
If such a postulate were restricted to standards of simultaneity, its realization
would result in expressing STR in the formalism of the so-called generalized
Galilean transformations (GGT).17 But to be consistent, one should also have to
correct the results of the influence of length contraction and time dilation on
physical standards of length and time units. This would mean regarding nothing else
but the familiar Galilean transformations as the 'true' kinematical transformations
holding in inertial frames.
FLoWOFTlME 63
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.
5. FLOW OF TIME AS ASELECTION RULE

Inconclusiveness of the argument from symmetry entails that, if anybody still
wanted to support his argumentation against objectivity of time flow on the theory
of relativity, he would have to make use of other properties of relativistic spacetime.
Such an alternative argument is indeed set forth. It utilizes the peculiarity of some
special solutions of Einstein equations, such as the so-called Gödel's cosmological
model. ls While in Minkowskian spacetime its symmetry causes the problem
of choosing one of many ways of separating time from spatial dimensions, in such
solutions no separation is possible in the global scale.
Yet, this argument can be reversed and launched at such solutions. For the
question arises if all solutions of Einstein equations are physical, i.e. are possible as
cosmological models. But their peculiarity lies in special distribution of matter and
thus they may be rejected only on the condition that the possibility of such
distribution can be questioned. Therefore, it is worth while to consider how to
satisfy this condition.
Einstein equations link spacetime metric to spacetime distribution of the energy-
momentum tensor, which appears here in the role of matter distribution. Actually
it is not the matter distribution itself, but at most its spacetime representation, since
matter is not extended spatio-temporally, but spatially. It is only the history of
a material object that is spatio-temporally extended. For a spatio-temporally
extended object to be the history of some portion of matter it must be possible to
reconstruct it as a temporal sequence of momentary states of a spatially extended
object. But this means endowing at least the spacetime regions that are filled with
matter with a structure that enables time flow.
What, then, about 'empty' regions? If they are to deserve the name of spacetime
regions, they must be 'places' for possible histories of material objects. It is true that
the presence of such objects would influence the spacetime geometry and thus,
strict1y speaking, their histories would not take place in the same regions. However,
these putative objects could be so small that their presence would not change the
geometry essentially. This justifies the postulate of possibility of extending over
them the time structure defined for 'filled' regions.
If the present line of reasoning is acceptable, then the possibility of introducing
global time should be regarded as a condition on which a four-dimensional space
64 JANCZERNIAWSKI
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.
6. GENERAL COV ARIANCE AND EXCESS OF GENERALITY

The point of view according to which the possibility of introducing global time
structure needed for enabling time flow should play the role of a selection rule
eliminating non-physical solutions of Einstein equations is justified by their
excessive generality. The latter follows from their generally covariant form, because
some transformations from the group of general covariance, interpreted actively,
enable generating from a physically admissible spacetime some evidently non-
physical structure. In the case of aspacetime of GTR with gravitation this fact,
expressed by the so-called hole argument,20 is obscured by the lack of simple metric
interpretation of spacetime coordinates. Thus, let us illuminate it on the example of
the spacetime of STR, considered as a special case of aspacetime of GTR.
In that spacetime the class of inertial coordinate systems is defined, in which
spatial coordinates are Cartesian coordinates, defined in the natural way on the basis
of measurements by rigid measuring rods in appropriate inertial reference frames.
In their turn, the time coordinates are defined by standard clocks resting in those
frames and synchronized on the basis of some procedure consistent with the
assumption of uniformity and isotropy of space. Any transition from one of such
coordinate systems to another, wh ich is realized by a transformation from the
Poincare group interpreted passively, corresponds to a change of reference frame
in which physical phenomena and the spacetime structure are described.
Of course, the spacetime can also be described in other coordinate systems.
Its description in such new coordinates can be obtained from the description in some
inertial system by a transformation from outside the Poincare group, interpreted
passively. However, the new coordinates will not, in general, have the same simple
metric interpretation as the ones of inertial coordinates. Thus, the corresponding
transformation not only does not realize simply a change of reference frame, but it
does not even have to be connected with such a change at all. Instead, it always
means a change of operational meaning of coordinates.
Let us consider a formulation of Einstein equations in some inertial coordinate
system. By a suitable transformation, their formulation in some other coordinates
may be obtained. Interpreted passively, the transformation means a mere change of
parametrization of the same spacetime. However, also the active interpretation is
possible, according to which new coordinates are defined in another structure,
obtained by a suitable transformation of the original spacetime. Thanks to general
covariance, the same Einstein equations hold also in new coordinates, thus,
according to the active interpretation, also in that new structure. Can we conclude,
therefore, that it is also aspacetime of GTR?
FLOWOFTIME 65
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
7. THE MEANING OF THE HOLE ARGUMENT

It might seem that, at least in aspacetime with gravitation, the distribution
of matter as its source should, by Einstein equations, univocally define the metric.
However, let us remember that not only the density of matter, but the field of its
velocity in a given spacetime region also depends on the metric. As a result, Einstein
equations mean not so much unilateral determination of metric by energy-
momentum tensor, but rather their mutual co-ordination. Thus, they express the link
between the metric and the energy-momentum tensor not only in the spacetime
before the transformation, but also in the anomalous spacetime after the
transformation.
Still worse, the restrictions imposed on the transformations from the general
covariance group are so liberal that they allow for transformations that differ from
the identity transformation only in a part of spacetime, e.g. in some restricted area
(the 'hole,).23 In the spacetime after such transformation physical phenomena
proceed anomalously only in that peculiar area, whereas they go on regularly
outside. It looks paradoxically especially in the case when that area is devoid of
matter, or at least there is so little matter there that the metric inside should be
practically wholly determined by the matter outside. Meanwhile, in spacetime after
the transformation a completely different metric in the anomalous region
corresponds to the same distribution of matter.
It is true that in the anomalous spacetime a new metric can be defined, relative to
which physical phenomena proceed regularly. Regarding it as physical realizes the
operation of geometrization of universal force. Thus, indeterminism in such
spacetime restricts itself to its non-physical aspects, lacking any operational
meaning. Yet, as we have already observed, the anomalous metric also satisfies
Einstein equations. Then, in principle a world should be possible in which the
phenomena would proceed regularly relative just to this very metric.
Such a world might have a very strange spacetime structure, especially when it
were generated by a transformation that would be non-trivial only inside the 'hole'.
No wonder that such solutions of Einstein equations are not taken seriously.
However, one has to formulate some criterion that would allow for their rejection,
which would not be exposed to the accusation of being ad hoc.
8. MACH'S PRINCIPLE OR REISTIC ONTOLOGY?

A proposed criterion of physicality of relativistic cosmological models consists in
regarding accordance with the Mach's principle as a selection rule for distinguishing
the physical solutions of Einstein's equations. 24 No doubt, it is a good idea. Yet, one
might wonder if such a condition is not too strong.
Full satisfaction of Mach' s principle will mean reducing the inertia of bodies
to free falling in the gravitational field generated by the totality of remaining bodies
in the Universe, while the influence of that field would have to be understood as
distant action. The conception of such interaction has evoked serious objections for
a long time. Thus, perhaps one should try to extract some weaker condition from
that of accordance with Mach's principle.
FLOWOFTIME 67
Mach's principle is based on a certain ontology. In its framework the world

is conceived not as a four-dimensional continuum of events, but rather as a system
of things, which move in space and retain their identity in time. Consequently,
spacetime is conceived as the geometrical structure not so much of the world, but
rather of the history of the world, whereas the world itself is extended only spatially.
It can be so only if the momentary states of the world are well-defined, so that their
unique sequence can make the history of the world. In other words, in spacetime an
absolute time structure must be defined, and that is, as we remember, a condition of
possibility of time flow.
From the above analysis it follows that if there is some soundness in the
condition of accordance with Mach's principle, then in any spacetime of GTR which
deserves being taken seriously there should be at least the possibility of introduction
of the structure enabling time flow. The above condition is weaker than Mach's
principle. Now, however, it is easy to see that such criterion would not be strong
enough to eliminate the above-mentioned anomalous spacetimes. Thus, its
strengthening proves necessary.
Taken abstractly, the notion of time flow leads to the paradoxes already noticed
by St. Augustine. 2S It acquires some meaning only in the context of the above-
mentioned ontology, which allows avoiding the identification of the world with its
momentary state, whereas that identification is the real source of those paradoxes.
A highly important assumption of this ontology concems the possibility of retaining
by a thing its strict identity through time. This assumption will prove very useful
with respect to the problem we are interested in.
From the perspective of this ontology, which can be called reistic, for the most
fundamental role is played in it by the concept of a thing - 'res' in Latin, the world
does not have any temporal extension, but only a spatial one. Presently it is in some
state, Le. it is qualified somehow. However, it has already passed through a
sequence of states in which it was and will pass through a sequence of states in
which it will be. Out of those two sequences, complemented by the present state, an
object can be constructed, which is extended also in time as an additional dimension.
The spacetime is its geometric structure. Thus, the eventistic ontology wh ich is
presupposed by the notion of spacetime is regarded by the reist not as literally true,
but only as some way of describing the history of the world.
Such an approach is supported by the results of reistic analysis of the notion
of event, 26 to which the eventists cannot oppose any analogous results of the
analysis of the notion of thing. Consequently, they cannot present any positive
evidence for the thesis that things do not exist (in the literal sense) as something
different from processes on the one hand, and from the momentary sections
of processes or states of things on the other. Moreover, no consistently eventistic
(i.e. one making no use of tools borrowed from reism) explanation of the fact, that
the world of our immediate experience is reistic, is known. Besides, it is evident that
the intuitions areist can rest on are supported by this fact.
68 JAN CZERNIA WSKI
9. REISM AND DYNAMICS

No doubt, for an eventist, such a tool borrowed from outside in order to account for
our ordinary experience is the idea of evolution,27 which evidently is a reistic
concept. In the process of biological evolution some things, namely living
organisms, undergo definite changes from generation to generation under the
press ure from other things which form their li fe conditions. On the other hand, it
cannot be seen how events could ever undergo such pressure from other events,
especially in aspacetime in which time does not flow and events, if they still
deserve this name, do not happen, but exist a-temporally.
Yet, the biological evolution is only a special case of state evolution of
a physical system. Such a system can always be regarded as consisting of things that
interact and influence the behavior of each other. Things are endowed with
properties, that is, with dispositions of two kinds. A reistic purist would say: they
are in many ways disposed in two senses of the latter term.
Dispositions of the first kind, which can be described as active, determine the
ability of a given thing to influence the properties and behavior of other things.
As examples, one might mention electric charge or gravitational mass,
i.e. gravitational charge, which deterrnine the ability of a physical object to generate
the appropriate field and thus to influence the motion of other objects. Next,
as examples of passive dispositions, determining the ability of a thing to be
influenced by other objects, one might mention the reciprocal of inertial mass or the
ratio of electric charge to inertial mass for charged particles.
Anyway, considering our purpose, concrete specification of active and passive
dispositions is not essential, and neither is the otherwise interesting problem of their
mutual dependence. What is important is that physical objects are endowed with
dispositions of both kinds, which entails that every system consisting of such
objects has the propensity28 to evolve in the direction that depends on them. In other
words, it has a certain dynamies, a special case of which is the above-mentioned
mechanism of biological evolution.
Of minor importance is also the question whether the direction of evolution
of a physical system is determined unequivocally or probabilistically, provided the
probability distribution is well-defined. What is important, is that the objects the
system consists of can retain their identity through time and thus also their dynamic
properties. Since, thanks to this, the propensities posessed by the system manifest
some natural regularity, described by the laws of nature. Apparently, dynamics,
without which constant regularities in nature are unintelligible, is a reistic notion,
toO. 29
Once again, it is not essential if this identity is retained strictly through time by
all things the system consists of. It is enough that within any sufficiently short
interval of time there are parts which preserve their identity. Their qualitative
changes, including as drastic ones as coming into existence or perishing, are not
accidental either, but result from the propensities of the system in its state that
precedes those changes.
Let us consider the flying arrow of the well-known paradox. Of course, contrary
to the opinion of Zeno, it is resting in no moment of its flight, since due to its inertia
FLOWOFTIME 69
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'.
10. COSMIC TIME AND TIME FLOW

There is still an objection against accepting the possibility of a reistic reconstruction
of spacetime as aselection rule in general relativity. WeIl, what might suggest itself
as the most natural candidate for the role of the absolute simultaneity as presupposed
by the idea of time flow, is the simultaneity in the sense of cosmic time. 30 However,
the latter is definite, i.e. unequivocally determined by the metric, only in such
cosmological models in which the distribution of matter satisfies some quite
unrealistic conditions. More realistic distribution results in the occurrence of local
singularities, which prevent extrapolating simultaneities defined locally, in reference
frames of fundamentalobservers, in such a way that one global simultaneity is
defined - one common for all such frames.
The assumption, however, that the 'true' absolute simultaneity must be
consistent with the simultaneities of fundamental observers, is not as obvious as it
might seem. Quite a contrary result is suggested by the construction of some
cosmological models. In particular, it is so in the case of Milne's model.
The construction of this model may be visualized as follows. 31 Let at the
beginning of the history of the Uni verse the whole matter be concentrated in a smalI,
in the cosmic sc ale practically pointlike, region of a three-dimensional Euclidean
space. Then an explosion takes place, and in its result different fragments of matter
acquire velocities of different values and directions. Assurne that the Universe
contains such a small quantity of matter, that gravitation can be neglected. Let us
also assurne that the matter behaves in accord with the relativistic mechanics. In
particular, it means that the velocity of none of its fragments reaches the velocity of
light and that a physical cIock connected with a given fragment of matter would
measure the proper time in its own reference frame.
The area in wh ich all the matter must be contained expands with the velocity of
light. In the passage of time the mutual distances grow between fragments of matter
endowed with different velocities. The more time has passed, the stricter the
70 JAN CZERNIAWSKI
distance of a given fragment from the place of explosion is proportional to its

velocity.
If with a fragment of matter a dock be connected which could measure somehow
the time since the moment of explosion, then that dock would determine some local
standard of time along its worldline. The docks connected with fragments with
different velocities would determine such standard in different spacetime regions.
Taken together, they would determine some global time in the whole spacetime,
which would be nothing else but the cosmic time, in the sense of the common time
of possible fundamental observers.
It might seem that in such a model the physical absolute rest is defined.
However, it does not have to be so, if appropriate statistic distribution of velocities
of fragments is assumed. Let us consider a hypersurface of a constant value of the
cosmic time. In a two-dimensional section containing one spatial dimension
it is represented by a hyperbola determined by the same value of the proper times
of inertial observers whose worldlines spread from one point.
Thus, all this spacelike hypersurface is a three-dimensional generalization of
a hyperboloid around the time axis in a resting reference frame. Yet, as defined by
the proper time, which is an invariant of the Lorentz transformations, it looks
exact1y the same in the frame of any other fundamental observer. In particular, it is
orthogonal to its worldline and tangential to the flat hypersurface of events
simultaneous in the observer's reference frame defined according to the standards
that hold in STR. Then, as matter in his neighbourhood behaves in accord with the
special principle of relativity, he cannot detect his motion on the basis of any local
experiment and, with appropriate global matter distribution, also on the basis of any
observation of a broader reach.
However, let us remember how the above model has been constructed.
According to that construction, global absolute rest is defined in it, although non-
physically, that is non-operationally. This means that the 'true', though also non-
physical, simultaneity, and not the simultaneity in the sense of the above-defined
cosmic time, is the simultaneity in the absolutely resting reference frame.
In the same way some cosmological models with gravitation can be
reconstrued,32 in which it also turns out that the 'true' time does not coincide with
the cosmic time. Thus, the intuition according to which in the spacetimes with a
cosmic time it is the latter that determines the structure corresponding to the time
flow proves false. This opens the possibility of introducing a non-physical absolute
time also in some spacetimes which lack cosmic time for the reason just considered.
Thus, the se1ection rule here proposed does not have to eliminate realistic
cosmological models. On the other hand, such a rule is supported by the positive
argumentation we have provided. So, in order to reject it, some new, sufficiently
strong argument should be found.
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.
11 Cf. Wald (1984), pp. 58-59; Anderson (1964), p. 36.

12 Cf. Earman, Norton (1987), pp. 518-519.
13 Reichenbach (1958), p. 33.
14 Ibid., p. 13.
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).
25 Cf. St. Augustine (1961), pp. 269-270.

26 Kotarbil'lski (1966), p. 430; cf. also Maxwell (1985), p. 29 and Czemiawski (1998a).
27 Cf. e.g. Butterfieid (1984), pp. 164, 166.
28 Cf. Popper (1991).
29 Cf. Maxwell (1985), pp. 31-35.
30 Cf. Czemiawski (l998b), pp. 152-155.
31 Cf. Rindler (1977), p. 206.
32 Ibid., p. 223.
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*
BRANCHING FOR A TRANSIENT TIME
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
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,
Section 4 defends this version of branching time theory against some popular
objections and illuminates a number of ontological assumptions behind SOBST.
2. THE FRAMEWORK OF SOBST

In this section I will use a standard method of logicians who account for modal
notions by producing a family of possible worlds or possible histories.
The technique of branching that is applied here is perhaps a less popular application
of this method. The name 'branching' goes back to early frameworks that accounted
for temporal ordering only. Possible histories shared then initial segments and were
arranged in tree-like structures. 'Initial' is no more a simple concept once relativistic
spacetime is considered, but it still remains true that histories in branching spacetime
have common segments. In contrast to most other modal frameworks, the models
which are presented here focus on 'little' things, which can be interpreted as events
and their outcomes. More precisely, once a modal structure is fixed by specifying
a nexus of histories, a net of events and outcomes emerges, and remaining analyses
can be carried out in terms of these notions. Significantly, given some algebraic
properties of outcomes, this feature permits applications to experimental data.
We begin the construction of a SOBST model by specifying a partially ordered
set W =(W; s). The elements of the nonempty set W are interpreted
as spatiotemporal points taken together with whatever is in them, Le., pointlike
concrete particulars. The relation x [y is interpreted as 'x is in the backward light
cone of y', or, 'x can causally influence y'. Standardly: x < y iff x::; y and x i= y. This
relation is then extended to subsets of W:
DEFINITION 1 (OF PRECEDENCE)

ForE,Fr;;; W,xeW(l)E-<x iff VeeEe<x;(2)E-<F iff VxeF E-<x.
E -< F means that the whole of event E is in the backward light cone of every point
from F.
As W allows for branching, two points can be separated not only

spatiotemporally, but also modally, when there is no single course of events that
comprise them both. The modally separated points will be characterized as upward
incompatible, with upward compatibility defined as below:
DEFINITION 2 (OF CaMPATIBILITY)

x,yeWare upward compatible iff there is a zeWwith z~x and z~y;
otherwise, they are called orthogonal (written x.L y).

Some special subsets of W will be called 'histories' (intuitively, possible courses
of events). Our definition builds upon an asymmetry between upward forks and
downward forks. An upward fork can be interpreted either spatiotemporally (two
points have a point in their common past) or modally (the points are upward
incompatible). To the contrary, downward forks permit a spatiotemporal
interpretation only, namely two points are in the past of a third point.
76 TOMASZ PLACEK
DEFINITION 3 (OF A HISTORY)

A subset h oJ W is a history iff his a maximal upward directed subset oJ W (i.e., Jor
all upward directed h' ~ W we have: h' ~ h implies h' = h ).
The set oJ all histories is denoted by H.
x
x y z
Figure 1: Upward and downward forks; the lines represent the ordering relation<.
This definition can lead to counterintuitive results, e.g., if our spatiotemporal

world comes to an end, the definition will treat our world as a huge number of
histories. The next section shows how to overcome this problem by building (W,S)
from spatiotemporal histories rather than carving histories out of (W,S) .
As histories branch, one can wonder where those things are located that are respon-
sible for some two points being upward incompatible. The concept that we need is
that of a set of splitting points for points.
DEFINITION 4 (OF SPLITTING POINTS FOR POINTS)

For any two orthogonal points x, y e W, we define the set oJ splitting points
C( x, y ) ~ W by putting z e C( x, y ) iff z is a maximal element in
{z e W : Z S x & z S y}. IJ x and y are not orthogonal, we put C( x, y ) = 0 .
To ensure that for any pair of orthogonal points x and y, C( x, y ) is non-empty

and that sets of splitting points behave 'nicely', we assume the following two
conditions:
(Cl) For any x,y,zeW, if x.lyand zSx,zSy, then there is some

tE C( x, Y ) with t ~ Z .
(C2) For any x,y,z,tEW, if X~Z and y~t, then C(x,y);;JC(Z,t).
Two further notions are required to introduce outcomes of events in W:

DEFINITION 5 (OF RELATIVE ORTHOGONALITY)

Elements x. y of Ware orthogonal relative to E, written xl.. EY , iff E -< x, E -< yand
C( x. y ) n E ~ 0 .
DEFINITION 6 (OF ORTHOGONAL COMPLEMENT)

For F ~ W • the orthogonal complement of F relative to E is the set Flog such that
xE Flog iff Vye Fxl..EY'
DEFINITION 7 (OF OUTCOME)

A subset F ofW is an outcome of E c W iff F = Floglo E
This definition ensures that an outcome of E is preceded by E and is located as

close
as possible to E. What the outcomes of E look like crucially depends on whether
and, if so, how many. histories split in E. Given the above definitions, the following
claims, proved in (Kowalski and Placek, 1999), hold:
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
DEFINITION 8 (OF EVENTS)

E c W is an event iff E ~ 0 and 3xe WE -< x .
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
PE( 1E ) =1 and for mutually disjoint ej E BE : PE( Ue = LPE( e

~ ~
j ) j ).
j;J j;J
In the other direction, on the basis of Theorem 2 we can represent classical

probability spaces in SOBST:
THEOREM 3 (REPRESENTATION THEOREM)

Given a probability space (B,p) with B complete and atomic, we can construct
a SOBST model that contains an event E such that (B,p) =(BE,PE).
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
outcomes of events. In particular, probability is not defined on combinations of

outcomes of different events. Any SOBST model has a collection of probability
spaces and at this stage there are no constraints whatsoever on how the probability
measures are related to one another. In what follows, we will put forward one mild
constraint on probability measure in SOBST models. The constraint requires that
probability measures JiE and JiF (corresponding to events E and F, respectively) be
related in some specific way to probability measure JiEuF (corresponding to the
union of events E and F). The requirement relies on the following two facts (proved
in (Placek, 2000a)):
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 .
FACT 2 If e and f are atomic outcomes of E and F, respectively, then e n f is either

empty or an atomic outcome of Eu F .
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
where the summation is over atomic outcomes only.

The right hand side equation follows from the fact that 1 F is equal to the sum of
all atomic outcomes of F; since these atoms are disjoint, the rest follows from the
definition of measure. The REQUIREMENT has many faces, depending on where E
and F are located with respect to one another.
Let us review the most salient cases. (1) If F -< E and Vhe'liE E h iff Feh,
then JiE( ej ) =JiEUF( ej nlF ) =JiEuF( ej ), which is a kind of Markov's property.
(2) Somewhat similarly, if E and F are contained in the same histories, and splitting
points are only in E but not in F, the formula above amounts to saying that as far as
80 TOMASZ PLACEK
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.
3. A GEOMETRICAL APPROACH TO BRANCHING

As already noted, Definition (3) can produce highly counterintuitive results. E.g.,
if our history comes to a close, the definition will regard it as a huge number
of histories. A similar problem emerges in modeling aspacetime structure in
a vicinity of a black hole (Belnap, 1992, p. 395). These problems can be avoided by
first relating histories to spacetimes, and then, on the basis of this relationship,
introducing other required notions. The essential part of this program is to define our
'causal ordering' ::; in terms of the topological structure of a given spacetime a.
A history will then be identified with aspacetime taken together with a full
specification of states of all its regions.
For a single Minkowski spacetime a , the definition of its causal partial ordering
::;". is straightforward.
DEFINITION 9 (MINKOWSKIAN ORDERING)

For lourpoints x and Y lram Minkowski spacetime a. we say that x<". Y iff x lies
within or on the backward light cone olY.
However, there is no uniform way of introducing a causal partial ordering for

spacetimes of general relativity, i.e. fourdimensional differential manifolds with
Lorentz's metric of signature +2. Some spacetimes of general relativity defy the
introduction of the relation sought after as they allow, for instance, for causalloops,
which imply a failure of antisymmetry. What one can do, however, is to delineate
a class of spacetimes of general relativity for which a partial 'causal' ordering is
definable. Yet, in this paper we will discuss Minkowski spacetimes only.
To model alternative possibilities, that is, alternative futures of events, we will
take a collection of Minkowski spacetimes and then paste some of them. Before
I say how such a pasting should be done, let me draw a distinction between two
notions related to branching spacetimes. While modeling chancy phenomena, one
may, on the one hand, be interested in points at which a split between alternative
futures is affected - these points I call after Belnap 'choice points' for spacetimes
and denote by CcrTJ the set of choice points for a given pair of spacetimes (1 and 17.
Some (idealized) chancy phenomena require a single choice point, like for instance
the decay of a pointlike particle in a given instant. Other phenomena involve a few
spacelike separated choice points - here EPR-like experiments come to mind.
It is natural to require that for any choice points x. y E CUI/' neither x is above y nor
vice versa. Given that a split is affected at a point x, any point y above x must be
already in one of the two alternative spacetimes. Thus, the alternative spacetimes are
already separated at this point y, and hence y is not a choice point for (1 and 17,
though it can be a choice point for some other pair of spacetimes. By the SOBST
reading of spacelikeness this means that any two choice points for any two histories
are spacelike. 3
However, apart from choice points, which play an active role in bringing about
the actual history, there also is another, more passive or geometrical concept to
consider. This concept answers the query of how a split affected at some choice
point(s) propagates globally. Or, to phrase it differently, on the supposition that two
spacetimes split at some choice point(s), what is the surface along which the
spacetimes branch? We need to know the answer to this query to properly paste
spacetimes. Let thus introduce here the geometrical concept of the suiface of
divergence D((1.17 )of spacetimes (1 and 17. If two spacetimes split at a single choice
point x, two things are intuitively clear: points 'below' x are shared by the
spacetimes involved, and no point 'above' xis shared by the two spacetimes (see the
left part of Figure 2). But what about the 'wings', that is, the totality of points that
are neither 'below' nor 'above' x? Here I fully endorse an argument of (Belnap,
1992, pp. 411-414) to the effect that the wings are shared by the two spacetimes.
To repeat this argument, suppose that some two points y and z are located in the
'wings', but do not belong to one and the same spacetime. An 'agent' responsible
for their being in two alternatives should be located 'below' y and 'below' z. But,
as far as our story goes, the only agent responsible for the split is the point x, which
is neither below y nor below z. Hence, y and z must be in the shared region of the
two spacetimes.
82 TOMASZ PLACEK
Having answered how a branching affected at a single choice point propagates

globally, let us next consider spacetimes that split at two spacelike separated choice
points x and y. In a case like this, depending on a frame of reference, there rnight be
three answers as to where these spacetimes split: (1) at x, (2) at y, and (3) at x and
at y. Using Belnap's strategy, one arrives at the surface of divergence schematically
depicted on the right-hand-side of Figure 2. The final point of this recipe for pasting
says that the surface of divergence should be shared by the spacetimes involved.
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 :
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.
(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
algorithm, as it associates outcomes of a corresponding event with results of

a quantum measurement and probabilities of corresponding outcomes of the event
with probabilities given by the quantum probabilistic algorithm. The SOBST
probabilities are then interpreted as objective weights of possibilities. It is not
known, however, how to represent the Schrödinger evolution of quantum states in
SOBST, or even the quantum states themselves. The framework sheds some light on
the notion of non-commutivity as it relates it to modal structures of branching
histories, but this does not in itself constitute an interpretation of quantum
mechanics.
(Q8) By assigning probabilities to outcomes rather than histories, does not
SOBST loose a chance to ground objective probabilities in the concept of proportion
of histories? This appears to be the objective of Storrs McCall's framework
of branching histories - see (McCall, 1994; McCall, 1997).
(R8) Technically speaking, SOBST allows for adefinition of the probability of
a history (Placek, 2000a) , but it makes this notion derivative with respect to
probabilities of outcomes. More philosophically, we need to distinguish two cases,
one with equiprobable histories and the other with histories having weights. The first
alternative, which appears to be assumed by McCall, is to me highly
counterintuitive. I treat branching seriously, so if I have grounds for believing that
two alternative outcomes of, say, a spin polarization measurement, have objective
probabilities 0.33 and 0.67, I take this as a fact that does not require positing any
further branching in the future whose only function is to yield these numbers.
McCall posits such further branching: in this example 33 histories in the first
outcome and 67 histories in the other outcome will do the work. Since in most cases
there is no evidence of corresponding chancy events, the postulation appears to be a
purely mathematical trick. Moreover, given irrational probabilities, branching must
be fairly complicated and involve infinite numbers of histories-see the prism
models of (McCall, 1997). Thus, if we believe that histories are equiprobable, we
face a dilemma between taking branching seriously and delivering probabilities by
means of proportions of histories. I decidedly take the first horn of the dilemma. On
the other hand, given that histories are assigned different weights of possibility, and
it is further explained how a given history becomes more probable as its earlier
segments get actualized, I believe that probabilities of histories will be derivable
from probabilities of outcomes of events, as they are in SOBST. Besides, I do not
see why explaining objective probabilities in terms of proportions of histories offers
a better foundation for such probabilities than accounting for them in terms of
weights of alternative outcomes of events.
(Q9) Apart from the rather weak REQUIREMENT, probability measures that
correspond to different events are not at all related. But the basic concept of theories
of probabilistic causality is that of one event making another event more (or less)
likely to happen. To accomodate this idea, doesn't one need to relate SOBST
probability measures somewhat more strongly? In particular, shouldn't SOBST be
made to mimic conditionalization of probabilities?
(R9) It appears to me that the idea is already incorporated in the SOBST
machinery. To illustrate, suppose that event A makes event B more likely to happen
than not. This is represented by putting B in the future of A (technically, B c JA) in
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?
To reply, we need to distinguish two ways of understanding someone's saying

that Humphrey could have had six fingers on his left hand. First, someone can say
this without believing that there was once a time at which it was indeterminate and
chancy whether or not Humphrey would grow a sixth finger on his left hand.
One asserts the statement on the basis of there being some human individuals, e.g.,
Ann Boleyn, who had six fingers on one hand, which is taken as the evidence that
Humphrey's having six fingers on his left hand would not contravene laws of nature.
This kind of possibility is what in branching frameworks is called conceivability.
And if Humphrey's having six fingers on the left hand is understood as a merely
conceivable state of affairs,we are not in any conflict with Lewis' framework since
to account for conceivability, SOBST assumes real Humphrey and his counterparts
in alternative histories, none of them overlapping with another. On the other hand,
the statement could mean that as a matter of fact, there was once a possibility that
Humphrey should grow six fingers on his left hand-this is called a possibility
based on reality.6 In this case, to think that Humphrey could have had six fingers on
his left hand is to think that at some earlier stage of his life, say, in his stage S before
Jan. I, 1913,7 the sixth finger could have grown on the left hand of that stage
of Humphrey's. To represent this, one builds a model with one event, Humphrey in
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.
There is an easyanswer to this in the SOBST framework. The crucial relation

of SOBST models is that of two spacetime points being orthogonal relative to an
event, written xl.EY and interpreted as saying that x and Y belong to different
histories that split in event E, and hence x and y cannot be causally connected. This
trivially yields this special demarcation principle: two pointlike individuals x and y
are part of the same world iff (1) for any event E, it is not the case that xl.EY and (2)
there is a z such that z Sx and z Sy. Clause (1) appears to incorporate the intuition
that Lewis voices. Clause (2) is rather technical, and is needed to account for
histories that do not have any segment in common. The principle easily extends to
other individuals.
The final, and 1 think most serious, of Lewis' arguments against branching
concerns the meaning of the future (Lewis, 1983, pp. 359-60):
( ... ) [A]n inhabitant of the shared segment cannot speak unequivocally of the world he
lives in. What if he says there will be a sea fight tomorrow, meaning of course to speak
of the future of his own world, and one of the two worlds he lives in has a sea fight the
next day and the other does not?
90 TOMASZ PLACEK
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.
But, as Lewis shows, none of these three options is tenable.

'The future holds a battle' is not a sentence you will normally hear. It is a
transcription in a non-tensed language of usual tensed sentences like 'There will be
a battle in the future', or 'There will be a battle tomorrow', or 'There will be a battle
next year'. SOBST, by supporting a tensed semantics, does not take Lewis' sentence
to be an adequate transcription of any of the tensed sentences exemplified above.
It moreover regards 'The future holds a battle' as a theoretical construct of non-
tensed semantics.Thus, it takes as a task the explication of the functioning of
sentences in future tense rather than Lewis' 'The future holds a battle'. Moreover,
taking the differences between the tenses as fundamental, there is little incentive to
believe that the explication of the discourse in future tense should be similar to the
explication of discourses in other tenses. In particular, since the assertion of 'There
will be a battle tomorrow' can mean as different things as my conviction that a battle
tomorrow is highly probable, or my intention to wage a battle tomorrow, or the fact
that the battle is necessary given the present state of affairs, some notions from
pragmatics are needed to account for assertions of sentences in future tense, apart
from a notion of truth.
Nevertheless, the basic concept needed for a SOBST account of tensed discourse
is that of outcomes of an event containing an act of assertion of a tensed sentence,
with words like 'tomorrow', 'next year', or 'future' indicating how large a part of an
outcome is meant. Thus, I reject Lewis' question of what 'the future' or
'tomorrow'stands for, by saying that on their own these words are not referring
expressions. Another notion needed is that of attrition, as with this we have the
notion of a once, but no longer, possible event, and the notion of the real history.
HRANCHING FOR A TRANSIENT TIME 91
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,
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)
Ifthere is Becoming, but

the physicists cannot know it, nobody can.
(Helena Eilstein)
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.
1. A LONG MARCH TOWARD THE MAIN TOPIC OF THIS ESSAY.
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
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
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
the view of the standard permanentism all do - Occur at definite spatiotemporal

locations and stand in definite (absolute or relativized to a coordinate system) temporal
relations to one another. Accordingly, in the specific language of permanentism there is
no place for expressions associated with the transientistic concepts of open and
extinguished possibility. With the elimination of these concepts the concept of act also
is rendered superfluous in the conceptual frame of permanentism.
The objective world simpJy ls; it does not happen.'
The difference between the Past, the Present and the Future amounts to nothing but an
admittedJy stubbom illusion" (A. Einstein).6
[1]f the experience of the Japse of time can exist without an objeetive Japse of time, no
reason ean be given why an objective Japse of time shouJd be assumed at aII.'
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.
2. THE DETENSER'S ENTERPRISE.

There are various names applied to the protagonists of the TIP dispute in the literature
of the topic. M. Capek used the name of etemists to refer to those who are calied
permanentists in my work. I do not use his terminology in order to avoid any illusion of
there being a kind of strict correlation between the fundamental standpoints represented
in the TIP dispute and the standpoints manifested in the dispute conceming problems of
temporal infinitude or finitude of the uni verse (in any meanings of the latter terms).
There also is a widespread custom to apply the names of detensers and tensers
- that is, derivatives of the term 'tense' (which in its ordinary use has onIy
a grarnmatical meaning) as means of reference, respectiveIy, to permanentists and
transientists. In accordance with that custom the very term 'tense', too, is often assigned
a new meaning, supplanting 'transienee foftime]' in a number of contexts. 9 The said
names are not seldom empIoyed as means of self-identification by representatives of the
respective parties. They occur in a variety of contexts where the topic under discussion
by no means pertains to tense as a characteristic trait of the grammatical structure of
a number of ethnic languages, particularly in our cultural domain. (To be sure, in
a number of other important contributions to the TIP debate, where the attention of the
protagonists is not captured by questions inspired by the occurrence of tensed
expressions in the said languages, these peculiar terms are not used at all).
I believe there are reasons not to endorse the above custom. It is unhelpful in
drawing a significant distinction among the protagonists of the TIP dispute. I mean
a distinction that should be made among them not with respect to their basic views on
the nature of time, but with respect to their attitude conceming the question what types
ofproblems should be put in thefoeus ofthe TIP debate and by what intelleetual means
that debate should be resolved.
AGAINST DETENSERS 99
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.
The belief in the affinity of common sense intuitions to permanentism is challenged by

the tensers. I do not doubt they are right about that: the common sense in our culture
(and, I presume, in any other human culture) is transientistic. 17 In particular, it
corresponds best to the possibilistic variety of transientism. Accordingly, the
interpretations given by detensers to the tensed sentence tokens used at the level of
common sense communication in our culture are bound to distort their meaning. There
arises, however, the question how much imponance should one assign to the above
state of affairs in our considerations about the temporal way of existing of the universe.
How should we approach the problem of the nature of time?
It is characteristic for detensers to believe in the need of a specific 'philosophical'
approach to the problems of the nature of time: an approach whose specijicity has to
consist in its independence of science.
[T]here are certain questions in the philosophy of time that science cannot shed any light
whatsoever... The attempt to infuse philosophicaJ issues of time with scientific ones
sometimes leads to unfortunate, because mistaken consequences... [E]ven if we ignore
scientific questions there still remain... many metaphysical and epistemological issues
conceming time. [The centraJ among them is] the status oftemporal becoming. 18
The transientist Q. Smith concurs with his adversary at this point:

The tensedltenseless issue [that is, the issue between transientism and permanentism] is a
matter not of physics but philosophy and it only can be resolved by the sorts of
considerations presented in this book ... [T]he correct interpretation of physicaJ theories
remains an open question until the metaphysicaJ issue regarding the tensed versus tenseless
theory is resolved by philosophers.1 9
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
Mellor maintains that

[although) the .. tensed sentences ... [do not) have the same meaning as the tenseless
sentences that give their tmth-conditions... those truth-condirions surely give their
meaning ... [A]nyone who knows that, for all dates T, 'it is now T' is tme during and only
during T knows what 'It is now ... ' means.23
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
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
4.THE COGNITIVE ATTITUDES OF THE PHILOSOPHISING ANIMAL.
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
agent acting in a given situation to themselves or to anybody else. As a matter of fact,

the concept of indispensability, as usually understood, is possibilistic too: the happening
of A is indispensable for the happening of B if and only if both A and B may happen or
fail to happen and B cannot happen in case A fails to happen. This concept should not
be identified with the perrnanentistic concept of preceding nomic necessary condition:
according to the conceptual frame of the permanentist an event of a type A Is a nornic
preceding necessary condition for a respective (appropriately Located in the spacetime)
event of a type B if and only if no event of the type B Occurs without being Preceded by
a respective (appropriately Located in spacetime) event of the type A.
All the above transientistic ideas are deeply entrenched in common sense.
Regardless of whether they are true or not; regardless of whether they can or cannot be
rendered plausible in the light of sophisticated cognition, it has to be acknowledged
that we all are in their grip when our minds are in the mode of 'practical Man'.
As I wrote in one of my works,29 the belief in the cognitive superiority of everyday
practice and common sense over the intellectual activities of theorizing and
philosophizing is a manifestation of the attitude that in my view may 'Je called
arrogance of common sense. I believe that the permanentist does not need, and his
adversary should not, submit to it. But the permanentist can consistently hold to his
position only if he commits hirnself to the view that there is an ineradicable
discrepancy between the attitudes of the theoretical Man in us and the attitudes of the
practical Man in uso
Man is a philosophizing animal. A philosophizing animaI is not able to be a
philosopher or a theoretician, free of the impact of emotions and involved in purely
cognitive research, for a!l time of its life. A philosophizing anima! is fOT most time of its
life an animal striving30 for survival (of its own, its offsprings and its community) and
fOT a bearable life. The consciousness of the theOTetical man in Man and the
consciousness of the practical Man in Man cannot be exactly the same unless the
practica! Man exerts violence over the theoretical Man and restricts his autonomy.
It should be noted that in my account of the above two aspects of the human nature
there is no hint at something that might be called 'cognitive complementarity' of the
two kinds of consciousness and no hint at something that rnight be caUed 'relativity of
truth'. I am, indeed, a supporter of the belief that from the point of view of cognitive
values the consciousness of the theoretical Man is superior to the consciousness of the
practical Man. It is just not the case that we can always be faithful to the verdicts of the
first type of consciousness in the practical life, when we display ourselves as cognitive-
evaluative-emotional-volitional subjects.
From the point of view of the permanentist, since the objective reality contains Man
among tl1e tota!ity of its components bound together in the nomic order of causal
relations, the 'stubbom illusions' which are characteristic fOT the philosophising anima!
in some situations of its life Must (in the sense of the regularity conception of
causation) Exist for the objective reality Beingjust the way it actuaUy Is. An illusion Is
an illusion, but Experiencing an illusion by a human individual Is a component of the
objective reality, and it belongs to the lattice of causal relations.
The adequacy of the above general description of 'the cognitive attitudes of the
philosophising anima!' does not depend on whether permanentism is true. Indeed, in
my book I used libertarianism to illustrate my point. 1 hope I managed to show there,
106 HELENA EILSTEIN
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.
5. BEYOND TIIE HORIZON OF TIIE Dff DISPUTE.

STILL FALSE STARTS.
The allegations 0/ absurdity. In the literature one encounters arguments by whose

means the representatives of one or the other party try to disclose either outright self-
contradictoriness in the viewpoint of their opponents or an inconsistency of that
viewpoint with some stipulations the opponents would not resign of.
Thus, e.g., in the writings of transientists one encounters the claim that in the
perrnanentistic uni verse all events would have to be simultaneous (in some privileged
frame of reference?). Milic Capek repeatedly expressed that queer idea in his writings. 31
As a matter of fact, his argument is directed against [prospective] determinism.
It is, however, meant to be deadly for perrnanentism, because according to Capek the
perrnanentist is logically bound to be adeterminist. Capek presents determinism in
classical physics as the conception according to which
the causa! relation was viewed as a relation of logical implication.... [E)very state of a
mechanica! system was regarded as logica/ly contained in its antecedent state... [lf) the
effect is implied by its cause, it should be simultaneous with it and any delay of its
happening is incomprehensible, since a delay is incompatible with the very nature of
implication.
This is an appalling confusion. Implication, of course, is a relation between

statements; it does not have any temporal connotations. It cannot obtain between an
effect and its cause (under any interpretation of the nature of the causal bond) and thus
it cannot 'by its very nature' necessitate simultaneity (in whatever meaning of this term)
of an effect with its cause. On the basis of the laws of some deterministic empirical
theory it rnay be possible to deduce that if at some time t (in the corresponding frame
of reference) an isolated physical system Is in astate of a given type, T, then (provided
the isolation holds) at adefinite later time, t', it Is in the state of some definite type T.
It surely cannot be deduced on the basis of such premises that the said later state,
considered with respect to its numerical identity would also have another location,
temporally closer or coinciding with what is taken in the respective argument to be the
initial state of the isolated system. Moreover, it may be the case that on the basis of the
laws of the same deterministic theory, and the same premise specifying the state
prevailing in a given physical system at time t, one can deduce that the concemed
physical system cannot be in astate of the type T' at any time be/ore t'. So much for
Capek's argument.
Permanentists sometimes also are inclined to recklessly charge their opponents with
absurdity. According to Earman and Gale,32
'The present' and 'now', like 'This time', are used to refer to a moment of time. Thus, to
say that the present shifts to later times entails that this very moment of time - the
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.
According to a number of transientists genuine events have to be conceived as

temporally pointlike, and thus the Now should be conceived as successively passing the
points of the temporal continuum. However other transientist, (as e.g. G. Whitrow37 )
believe that processes consist of discrete events which despite their temporal
extendedness do happen by way of unanalyzable, all-at-once jumps. (Whitrow
conceives time itself as consisting of discrete 'chronons' of nonzero extendedness; note
that this idea is not implied by the one of discreteness of events).
Both the above conceptions - of pointlike as weil as of temporally extended
events - are internally consistent and compatible with both transientism and
permanentism. Note, however, that against the background of transientism both these
versions imply that the happenings of events cannot be 'directly' perceived.
If ternporally pointlike events exist, we cannot observe them; and if genuine events are
temporally extended, we also cannot 'directly' observe the discontinuity in the
developrnent of processes, for these events surely have to be microscopic. What we
observe turns out to be a delusive picture of the 'flow of time', one analogous to the
delusive picture of water as a continuous medium.
M. Capek38 rejects both the above transientistic approaches. He insists on the
genuine continuity of the flow of time which according to hirn should be conceived as
incompatible with the existence of either instants (-individuals in the Cantorean
temporal continuum) or chronons; it also should be incompatible with the analyzability
of processes into successions of particular events. In his view the 'natural' continuurn
should not be conceived as intrinsically 'pulverized' into elements.
The Cantorean approach to continuum as to a set, of the power c, of its pointlike
elements, has been repeatedly contested in favor of the search for a more holistic
approach to its nature. In some cases that was done for reasons pertaining to philosophy
of mathematics itself. In other cases, however, that was done due to the conviction of
some thinkers that the Cantorean approach belies the nature of the real, physical,
spatiotemporal continuum. Authors as outstanding as H. Poincare and H. Weyl (not to
mention pure mathematicians as L. Brouwer and A. Heyting) participated in these
debates, for whose presentation there is no place here. But in his paper in this volume
(as weil as in a number of his more special works) M. Heller presents the most recent
reasons, inspired by contemporary physics, for turning to such a more holistic approach,
and points out that a consistent theory of a 'non-Cantorean' continuum is provided in
contemporary mathematics. In noncommutative geometries one does not have to do
with the concepts of a point and its neighborhood; the respective spaces receive only a
global characteristic. And an appeal to a noncommutative geometry may be needed in
the quantum theory of gravitation.
To be sure, localization [of physical entities I in spaee and time eonstitutes a very basic
element of our everyday experienee. Therefore only those physieal theories whieh bring
forth a model of the uni verse in the smallest seale ean resort to a noneommutative
geometry ... That seale is rendered by the so-called Planek's threshold that eorresponds to the
distance whieh is equal to about 10-33 em and to the temporal distanee whieh is equal to
about 10-44 s. Beyond that seale the spaee and time have to be attributed their usual 'Ioca!'
properties. A physical theory whieh deseribes the fundamental level must thus ... allow the
points and their neighborhoods to 'emerge' from the noneommutative properties [of
spaeetime I after transcending the Planek threshold.39
110 HELENA EILSTEIN
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.
6. ON THE TRANSIENTISTIC ARGUMENT FROM INDETERMINISM.

It is typical for possibilistic transientists to believe that their theory of time acquires a
substantial support from indeterminism of Quantum Mechanics. They perceive a direct
link between indeterminism and possibilism. Let us dweIl on this matter here.
The victory of indeterminism in the domain of microphysics has not been
irrevocably established. Notably the standard interpretation of Quantum Mechanics
(which is indeterministic) is beset with very serious conceptual difficulties. A number
of its keenly discussed alternatives also have an indeterministic character. However,
one encounters among them also the deterministic alternative developed by D. Bohm.
The very unintuitive, but attracting a number of supporters, 'many-worlds
interpretation' (and its modification, the 'many minds' interpretation) initiated by the
works of H. Everett and B. DeWitt also should be considered deterministic, albeit in a
peculiar way, since it defies the conception of possibilities which may fail to actualize.
In this essay, however, I do not need to dweIl on that dispute, because the proton
pseudos of the above transientistic argument is independent of the truth-value of the
thesis of indeterminism.
Indeed, an actually viable response of the pennanentist must be based on the
analysis of the logical relation between prospective indeterminism (to be called
indeterminism in the present context) and possibilism.
Indeterminism is nothing but the denial of prospective determinism (to be called

determinism in the present context). It is thus best to start the analysis of indeterminism
with the analysis of the metaphysical hypothesis of determinism. In order to grasp the
above named logical relation it is sufficient to deal with determinism and indeterminism
against the background of Newtonian spacetime. It is thus sufficient to assume here, for
the sake of argumentation, that spacetime is stratified in a unique way into genuine
instants and that it is endowed with a global arrow of time, where the concept of that
arrow should be assumed to be applicable with reference to all structural levels of the
uni verse. It also is best to resort to the specific language of possibilism in order to
express the thesis of determinism.
Determinism, considered with respect to all the above underlying presuppositions, is
aversion of actualism. It is a metaphysical hypothesis accordin~ to which every act has
its nomic sufficient condition (in the sense of regularity theory 2) for its Occurrence at
every instant which Precedes its Occurrence. Every act constitutes the last link in a
continuous chain 0/ acts which Precede it and each of which is a nomic sufficient
condition/or alt the Succeeding ones. Accordingly, in a deterministic conceptual frame
there may or may be not a pi ace for the distinction between an act in the Future and a
fact, but there is no place there for the concept of an open possibility which may or may
not actualize.
Determinism, as it is seen from the above, does not reduce to actualism; hence the
denial of determinism does not imply the denial of actualism. For indeed, determinism
involves the stipulation that some respective relations of nomic sufficient conditioning
obtainbetween every particular event and some respective events or sets of events
which Precede the event in question. That stipulation goes beyond bare actualism.
It follows from the above that since indeterminism is nothing but the denial of the
general thesis of determinism, it does not, in its turn, presuppose or imply possibilism.
In order to express the general thesis of indeterminism one does not need to attach any
objective significance to the concepts 0/ open possibilities, their actualizations and
extinctions. While possibilism is indeterministic, indeterminism in itself is consistent
with actualism; it also is compatible with permanentism - which, as it has been
pointed out above in my essay, is neither presupposed nor implied by actualism but
implies it. Indeed, in contradistinction to Einstein, whose concept of the universe was
both permanentistic and deterministic, a number of contemporary permanentists are
entirely reconciled with indeterminism, particularly in the quantum domain.
An analogy may help to grasp the difference between indeterminism and
possibilism. Let us consider, just for the sake of that analogy, a metaphysical hypothesis
to which the name of transspatial determinism will be assigned here. In order to
introduce that idea let us first of all adopt here, for the sake of argument, the classical
transientistic approach to time. Assume that we employ a frame of reference where the
time coordinates label the genuine instants of time - the ones that correspond to the
successive Nows. Under the above presupposition the doctrine of transspatial
determinism would say the following: For any instant of time, t, representing the
current Now, consider the state of affairs prevailing at it inside an enclosure of a spatial
area, which may be arbitrarily small. That state of affairs comprises nornic sufficient
conditions of happening for every event that is happening just at that instant anywhere
in the universe. 4 (That is, one may say, the enclosed area objectively contains infallible
112 HELENA EILSTEIN
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
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.
7. THE QUESTION OF A LADDER AND THE QUESTION OF A CUMBER.
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
In his paper 'Cosmological Singularity and the Creation of the Universe,47

M. Heller writes with respect to one of the relativistically adrnissible model of the
114 HELENA EILSTEIN
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
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
possibilistic transientism is concemed, those 01 being possible or being actual, or being

an extinguished possibility - represent ways 01 existing 01 particular events at different
tirnes; they, however, do not characterize events relative to anything. In particular,
events are (when they are) in these modi neither relative to particular stratijications 01
spacetirne into instants, nor relative to ontological modi olother events. Accordingly,
the Present, if conceived as the transitory instantaneous spatially ubiquitous
hypersurface, which forms the shifting closure between the Past and the Future, cannot
be relativized to any stratijication 01 spacetime into instants. On the contrary, the shift
of the thus conceived Now has to correspond to the stratification of the totality of events
into the classes of Copresentness, that is, of genuine Simultaneity. The attempts, thus, to
reconcile transientism with the standard interpretation of Special Relativity by means of
direct relativization of ontological statuses of events to particular stratifications of the
spacetime into instants must be considered a grievous violation of the meanings of the
terms in question in the genuine conceptual frame of transientism. And that is exactly
what the most naive 'reconciliators' try to do. The supposition that an event may be
'still' merely possible from the point of view of some frame of reference while being
'already' actual or no more possible with respect to some other frame of reference
appears particularly incomprehensible considering the fact that to call an event merely
possible means to avow that it may or may not actualize.
It does not, however, help the 'reconcilliators' when they turn to a more
sophisticated approach, consisting in giving up the idea of instantaneous spatially
ubiquitous Now and stipulating that Special Relativity admits only spatiotemporally
pointlike Here-Nows, which are supposed to run along every particular worldline and
successively coincide with particular fourpoints. If one additionally supposes that for all
worldlines which intersect at a given fourpoint the Nows running along them 'reach'
that point concurrently, dividing every of these lines into its past and future parts, then,
according to this approach, nowness can be attributed (at the appropriate time) to
particular lourpoints, that is, to sets of spatiotemporally coinciding Minkowskian
events; so that every actual Minkowskian event has to be 'present at some time' .51 That
conception also leads to the relativization of ontological states of events and thus cannot
be reconciled with the stipulations of transientism. All above discussed attempts to
reconcile transientism with the standard interpretation of Special Relativity are thus
doomed to failure. I tried to show it as clearly as I was able to in my above mentioned
polernics with Professor Shimony. Actually, however, the matter has been clarified long
ago, e.g. in the papers of H. Putnam and C.W. Rietdijk52
As I pointed out in my polernics with Shimony, supposing that a given fourpoint, P,
is now, one has to stipulate that events Located at or inside its forward light cone belong
to the Future and events Located at or inside its backward light cone belong to the Past.
Events outside these both light cones are supposed to be in the state of a peculiar
ontological limbo, being neither past nor future, nor present in the traditional sense,
since, e.g., the relation of absolute temporal Precedence prevails between some of them.
I suppose, however, that from the point of view of transientism most worrying is not the
appearance of the idea of an ontological status not known to the prerelativistic version
of that doctrine, but the peculiar correlation of the ontological states of all Existing
events to the ontological status of a particular fourpoint, P, which is in no way
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
'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
strongest incentive to challenge the standard interpretation of Special Relativity comes

nowadays, as far as I know, from reflections on Quantum Mechanics. Notably the most
intriguing aspect of Quantum Mechanics is its nonlocality, as manifested in correlations
discussed in the EPR 'paradox' or in the Bohm-Aharonov effect - and indeed as
suggested by the very concept of the 'instantaneous' collapse of the wave packet.
In that context some of the proposed interpretations of Quantum Mechanics (like that
of D. Bohm and BJ. Hiley or that of Ghirardi, Rimini and Weber) invite us to come
back to the idea of absolute simultaneitli In his paper 'Quantum Theory without
Observers,63 Sheldon Goldstein points out that none of the theories sketched there as
the most serious rivals to the Copenhagen interpretation is Lorentz invariant;
whereupon he expresses a 'belief that 'such a theory is possible' .
In the view of the above mentioned challenges to the relativistic denial of absolute
simultaneity the permanentist faces the question whether these challenges should also
be considered menacing to his own point of view. I will deal with this question soon, in
a broader perspective.
Let us, narnely, take it into account that despite its great importance for
contemporary physics and its practical applications Special Relativity is nothing but an
approximation to General Relativity. Thus the dispute concerning the existence of a
privileged stratification seems to require a consideration of this problem in the light of
General Relativity.
In contemporary cosmology an important role is played by a model of the universe
as a highly homogeneous and isotropie liquid, with clusters of galaxies as its 'atoms'.
Working with this model one indeed is able to introduce the so-called 'comoving'
coordinates by whose means one reintroduces the 'cosmic time', a relativistic
counterpart to the classical idea of absolute simultaneity. While according to some
physicists the above considerations reconcile General Relativity with the concept of
absolute simultaneity, other ones view the model admitting 'comoving coordinates' as
nothing more than an idealized presentation of the real uni verse and consider the
reintroduction of the idea of an absolutely privileged stratification unwarranted. 65
These observations allow us to pass over to the main point of the present section of
my essay. Unfortunately for the transientists, no attempts to reintroduce a privileged
stratijication of spacetime into physics can secure victory for their doctrine. The reason,
to put it metaphorically, is that 'the existence of a ladder does not warrant the existence
of a climber'. The existence of a physically privileged stratijication of spacetime into
spatially ubiquitous instants would not suffice to warrant the objectivity of the current
Now. As it is clearly presented by Saunders,66 the existence of a privileged 'foliation'
would furnish only a necessary condition, but by no means a sufficient condition for the
victory of classical transientism. Permanentism is entirely compatible with the idea that
these or other nomic or de facto features of the uni verse provide for a privileged
stratification of spacetime into instants.
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
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
meaning than in the combination 'T/P'.

122 HELENA EILSTEIN
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
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.
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
ON EXISTENCE OF THE FUTURE
Abstract. What is the future? Does the future exist?

These questions play an essential role in disputes on some crucial philosophical issues, such as the
problem of determinism and free will as weil as problems conceming validity of fundamental principles
oflogic.
111 order to solve these questions one has to distinguish three different modes of existence of
objects which belong to the temporal sphere of being. They are: the real (actual, present) existence, the
postreal (past) existence and the potential (prereal) existence. The notion of potential existence rnay be
supposed to apply to two categories of objects. The first one comprises objects which in fact will become
real. They may be referred to as potential objects (sensu stricto). The second category consists of quasi-
potential objects which never will come into existence.
This paper brings forth arguments for the thesis that the future is the domain of objecsts which are
potential sensu stricto. Only this definition presents the future as an ontologically homogeneous domain.
And only on its basis the language systems which apply to the temporal sphere of being can satisfy the
principles of non-contradiction and of excluded middle.
1. THE PROBLEM OF EXISTENCE OF THE FUTURE
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.
(A). An Examination ofthe Positive Answer (the Thesis A).
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
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.
(B). An Examination ofthe Negative Answer (The Thesis B).
There are some arguments for rejecting also this thesis.

Fin,tly, it runs contrary to the common view (which also has rarely been brought
into question in philosophy), that there are three domains of time. On the assumption
that one of these domains does not exist, the very existence of time also comes into
question.
Secondly, assenting to the above thesis makes it difficult to understand time, as it
is usually done, as a continuum of the same structural type as the one-dimensional
space of real numbers.
Thirdly, if the future did not exist, then the range of the basic temporal relation,
namely that of being earlier, would be limited to the domains of the past and the
present. Consequently, it would be groundless to assert that past and present objects
temporally precede the future ones. It is, however, intuitively obvious that e.g. every
presently occurring event precedes in time everything that will occur in the XXII
century.
Finally, the thesis (B) requires one to accept a very questionable view on the
truth-values of statements which are intentionally considered to refer to future states
of things. As a matter of fact, in view of this thesis the very meaningfulness of such
statements becomes questionable on account of their referring to things whose
existence has to be denied as a matter of principle.
In the light of all the above arguments it seems impossible to resolve the problem
of existence of the future by simply giving a 'yes' or 'no' answer to it. As I see it,
in order to be able to deal with this question one has to introduce certain distinctions
within the general concept of existence. A possibility of granting truth to some
appropriate modifications of both contending theses, (A) and (B), may be provided
by means of distinguishing different modes of existence, as specific for different
categories of temporal objects.
In order to develop this idea it is necessary to consider whether in the light of
ECE existence can be granted to both the other temporal domains of being. With
this purpose we have to express ECE a bit more accurately. Let us thus accept the
following assertion:
Existence of an object x is granted in virtue of ECE, if:
(1): x belongs to the material sphere of being;
(2) x satisfies at least one of the following conditions:
(2 ') a present direct or indirect observation (an observation being performed
'now') of xis possible;
(2 ") a present account (an account rendered 'now') of a possible observation of
x is possible.
The above formulation of ECE may raise various kinds of doubts. There seems
to be a need to specify more closely the meaning of some terms employed in it and
EXISTENCE OF THE fuTURE 129
to make explicit some of its underlying assumptions. I have to restrict myself to a

brief discussion of some of the corresponding issues.
The above formulation represents an approximate, simplified version of ECE.
One of the simplifying assumptions consists in not taking into account that all
physical processes, and in particular those wh ich constitute indispensable elements
of acts of observation, have a limited speed of propagation. Disregarding this fact
amounts to the assumption that an act of observation and the observed object may
be simultaneous ( to be more precise, that an act of observation may be simultaneous
with the observed state of the object in question). In a great number of cases,
however, the above assumption can be admitted as an innocuous idealization and,
consequently, one also is allowed to disregard the unintuitive condusion that
present objects never are directly observable.
To proceed, in the above formulation of ECE it was indispensable to resort to
the concepts of possible observation and possible account on an observation.
These concepts are not quite dear. In order, however, to make them precise it would
be necessary to delve into many controversial issues. This is why I have to resort
here to a philosophical fiction with the purpose to elicit in the reader an intuitive
understanding of the intended meanings of the terms in question. I am going to
employ some components of the well-known fable of the Laplace' an Demon.
Let us assurne that there exists an eternal conscious Being - 'the Universal
Observer' - endowed with a sensory apparatus which at every instant of time
makes hirn able to directly observe everything what is going on in the material world
at that very instant. Let us also assurne that this Being is able to permanently
preserve in his memory an absolutely precise picture of whatever was observed by
hirn.
Our Universal Observer still is supposed to be subject to some basic limitations.
He cannot observe any past or future things.
I would like to stress that the concept of the Universal Ob server is not attributed
here any metaphysical or theological significance. I believe he is a fictitious
personage which is introduced here just for the above said purpose. With that
purpose in mind let us adopt the following conditions:
(i) Every observation which is being performed by the Universal Observer is
a possible observation;
(ii) The image of an observed object, state of affairs, etc., in the memory of the
Universal Observer is a possible account (record) of an observation.
Against this basis ECE allows us to acknowledge the existence of two dasses of
objects. Every object presently 'given' in a possible observation belongs to the first
of them. These are the objects which satisfy the above condition (i). Assuming that
the Universal Observer creates records of his observations at the very instant of
performing them, the objects in question also satisfy the condition (ii). It is natural to
refer to these objects as to 'the present ones' - 'the ones which exist now'. And
since it seems obvious that the dass of such objects is ne ver empty, one can state
that ECE allows us to acknowledge the existence of presently existing objects and
thus the existence of the present.
That, of course, is not the only reason we can give in support of the thesis that
the present exists. On the contrary, one can legitimately hold that there is not even a
l30 STEFAN SNIHUR
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
2. THE PROBLEM OF THE ONTOLOGICAL STATUS OF THE FUTURE

Among those who accept the thesis (A') there still may obtain profound differences
in their views concerning the character of the future. The point is that according to
the ordinary understanding one may speak about two very different categories of
objects to which potential existence should be gran ted. The first of them comprises
objects which indeed will become real in the course of time. Of these objects only
it is legitimate to say that they do not yet exist really. The other category is supposed
to comprise objects which will never acquire the status of real existence. The
question arises whether objects of that category should be gran ted any mode of
existence? I will deal with this problem in the later course of this essay.
It is useful to introduce terms by whose means one could distinguish the above
two categories. Let us refer to objects (or putative objects) of the last category as to
the 'quasi-potential' ones and reserve the name of 'potential objects' to those of the
first category.
And now let us examine the following alternative answers to the question what is
the future:
(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 widely understood material sphere of being. The terrninology of temporal

languages of course has to be adjusted to the task of description of various kinds of
temporal relations and such dependencies, which presuppose temporal relations.
Only in such languages it is possible to formulate statements abut the future.
The two above factors are of essentially different nature. We can influence the
first of them - that is, we are able to modify our language, improve its terminology,
search for more adequate means to represent our ideas. The second factor, however,
is completely independent of uso For indeed it does not depend of us what is the
nature and the ontological status of the domain of being to which we are directing
our attention and whose image we are trying to create by means of the conceptual
apparatus of a given language. We have no influence on the mode or modes of
existing of the corresponding objects.
Let us now assume that J is a temporal language, all of whose sentences which
concern the temporal domain of being refer to such events, situations, states of
things etc which are unambiguously distinguished and are strictly 'located' in time.
In general, let us assume that there are no obstacles due to the first factor for the
satisfaction of the above logical principles by sentences of J. Let Z be a sentence in
J, which states that a certain event S is '/ocated' at a certain instant m. In addition
let us assume that the present instant is earlier than m; but let us also assume that Z
contains no temporal occasionals ( it does not specify whether m is a past, present
or future instant.) It is thus a sentence, which falls into the category discussed in
(H). Finally let us assume that the previously discussed thesis B (denying any mode
of existence to the future) is true. In view of these assumptions it is presently the
case, that there does not exists, in any sense, an object (event) to which Z might
refer. Therefore presently Z must satisfy one of the following alternatives:
(a) Z does not have any truth-value.
(b) Z is false.
(c) Z is possessed of a truth-value which is different of either Truth or
Falsehood.
And now let assume that S indeed is going to happen at the instant m. Z will
then become true. In other words, whatever is the present truth-value of Z, it will
subsequently change. That means that J does not satisfy the condition (2) of the
assumption (H). The principles of non-contradiction and of excluded middle are not
valid for all its sentences.
It follows from the above that existence of all three domains of time is
indispensable for the validity of the basic principles of logic in a temporallanguage.
This necessary condition, however, is too weak in order to provide us with a
criterion for the adequate choice between the theses (Tl) and (12). Indeed, it does
not warrant that every sentence in a given temporal language has a permanent truth-
value unless it contains temporal occasionals. For it is not incompatible with the
assumption that among the components of the object domain of such a language
there are objects wh ich fail to permanently exist in the absolute sense (while their
mode of existence may change) - that is, objects which do not exist in any way
just at certain moments or periods.
Let a be an object of such a kind in the object domain of J. Assume that a exists
at the moment m., but not at the moment m'. Under that supposition it is possible at
136 STEFAN SNIHUR
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
THE PROBLEM OF TIME IN THE PHILOSOPHY

OF ROMAN INGARDEN
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.
1. INGARDEN'S THEORETICAL STANCE

In order to understand Ingarden' s views on time and their place in his philosophy,
one must take account of the point of departure of his endeavour and the direction in
which he wanted to move. Ingarden was an outstanding - and, perhaps, the most
faithful - disciple of Edmund Husserl's from his early, Göttingen period.
Nevertheless, he could not accept the idealistic tendency of his master's thought.
Consequently, he has set as the main task of his long and fruitful theoretical activity
to substantiate the thesis that the 'external' world of things, persons etc. does indeed
exist independently of the minds of people who experience it - not in some subtly
philosophical, Pickwickian sense, but in a sense which basically accords with our
common intuitions.
In this context, the problem of time, central for any philosophy which has the
ambition of trying to understand and explain the sense of the world and of our life,
became particularly burning. Namely, Husserl had a tendency to derive the whole
sense of reality from its 'constitution' in the 'strearning' human consciousness.
It is clear that even the fact that he regarded the final source of this constitution -
the 'primarily constituting absolute, transcendental subjectivity' - as not temporal
does not diminish the central role of the analysis of time in his philosophy.
2. TIME IN THE PARIS LECTURE FROM 1937
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.
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
In this part Ingarden also discusses some theoretical consequences of accepting

the second way of experiencing time.
INGARDEN ON THE PROBLEM OF TIME 139
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.
3. THE ANALYSIS OF TIME IN THE LAST SECTIONS OF 'MAN AND TIME'
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
4. TIME IN DER STREIT ÜBER DIE EXISTENZ DER WELT

The mentioned fundamental difference between two conceptions of time is also
described by Ingarden as the difference of opinion concerning the sort of objects
which, in the last res ort, exist independently of other things - processes (or, we
could add, one process encompassing everything that exists) or objects enduring in
time. In Chapters V and VI (in the first edition Ch. VI and VII) of the first volume of
Streit lO Ingarden speaks about the problem of the existence of temporal objects -
events, processes and objects enduring in time; in particular, among the last
category, of living individuals and human persons.
In the first place, in accordance with the results achieved in 'Man and Time', he
states that only the first view of time, called by us 'realistic', enables us to properly
grasp the idea of a process and, all the more, of the objects enduring in time.
Moreover, the existence of objects enduring in time is already implied by the
existence of processes which need those objects as their ontic basis. We may call
this thesis aformal argument for the first view of time. It is broadly discussed in the
formal and existential analyses of the first two volumes of Streit. It joins the
argumentfrom humanfreedom outlined in the last paragraphs of 'Man and Time'.
The argument from freedom is c10sely related to Ingarden's considerations in his
essay on responsibility where he states that responsibility is linked with time in three
respects; namely, in order that responsibility could take place
1. the author of the deed must remain responsible after its performance,
2. the values for whose destruction the author may eventually be responsible
must remain to be valid.
INGARDEN ON TUE PROBLEM OF TIME 141
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.
5. TIME AND 'CONSTITUTION'

'Constitution' was one of the basic concepts and problems of the philosophy
of Husserl. But, as Ingarden remarks, Husserl has never explained this concept in
a satisfactory way - at least in his published workS. 12
Ingarden discems three main meanings of the term 'constitution' in Husserl:
• Adetermination 0/ the sense of an object by a multiplicity of aspects
(appearances) and intentions of acts. Those intentions and aspects are regarded as
conditioning the possibility of precisely such a sense being given to uso They are
also treated as, so to speak, ready-made, and therefore this is a static conception
of constitution.
• The process of creation of such a sense and of its transformations.
• The growth of primary data and of mental units in the succession of phases
of the primary flow of consciousness in the Husserlian 'absolute subjectivity'.
There are, says Ingarden, two possible extreme interpretations of 'constitution'
in this last and, for Husserl, basic sense. In the first interpretation (a) which he
approves as the philosophically right one, amental unit (Erlebnis) develops and
realizes itself owing to its own dynamics of existence and growth. According to the
other understanding (b) which is suggested by the use by Husserl of the terms
erzeugen and schaffen in the description of the function of constitution, immediate
memory (retention) and immediate anticipation (protention) would create the
mental unit (experience) and solely the primary, 'flowing' consciousness would not
be created but would grow by itself, phase after phase.
Those two possibilities - (b) all the 'constituted' units being created by the
creative intentions of conscious acts and (a) the actual growth of them according to
their own essential laws (as stipulates Ingarden) - are closely connected with two
interpretations of that what, for Husserl, exists 'absolutely'. That is to say:
corresponding to (b) - it is only the existence of the subjective stream of
consciousness, and to (a) it is more, namely the absolute existence of any immanent
mental unit. This last view Ingarden attributes to Husserl's Ideas.
Let us now reflect on the two ways of experiencing time described by Ingarden.
The first, 'realist' one, deals with time in the context of man in his real environment;
it is, so to speak, the time of the world in which man lives, the time of things and
persons in this world. Moreover, Ingarden stresses the aspect of acting and says
clearly that time belongs to the mode of being of the very object and is not a form
superimposed from outside;13 he also clearly refuses 'to approve the conception of
142 ANDRZEI POLTAWSKI
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,
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?
6. TIME AND 'PURE CONSCIOUSNESS'

At the end of the second volume of Streit Ingarden wrote:
Pure consciousness seems to be in concreto enclosed in the innermost core of the 'real
self and may be as such isolated from this self only to a certain degree and purely
abstractively, so to speak only mentaJly. If it is so, then it exists as a peculiar element
within the real world and cannot be tom out from the whole net of causal relations;
while, just as 'pure' constituing consciousness it would have to remain outside the
world and, in particular, to remain free from any involvement into the net of causal
relations in it.. [... ] This apartness and delimitation [lngarden seems to want to say: this
absence of apartness ... ] is not a question of the transcendental method and, in general,
not only a question of method which we would have to follow more or less strictly. It is
rather the strict connection between the 'pure' and the 'real' self which. does not allow
us to ins ist on the strict application of this method. 20
Treating 'pure consciousness' as an element of the real world, of a real human

person Ingarden would have to agree, that its time is identical with the time of this
real person and, consequently, that there does not exist any particular time of pure
consciousness. But let us ask: would he also accept the thesis of the impossibility of
pure phenomenology of time, of its 'pure' intuitive apprehension? Placing pure
consciousness in the real person does seem to make the acceptance of such an
apprehension difficult. But we mayaiso ask whether Ingarden's analyses do not lead
in the direction of abandoning the concept of 'pure consciousness' altogether.
Following the Cartesian construction of 'cogitare' as the sphere of indubitable
knowledge, Husserl affirmed that pure consciousness :
1. is indubitably given in immanent perception - at least in its existence and its
essential structure,
2. 'in its "purity" must be reckoned as a self-contained system of Being, as
a system of Absolute Being, into which nothing can penetrate, and from which
nothing can escape' ,21
3. 'in princi~le nulla "re" indiget ad existendum' - does not need anything
in order to exist, 2
4. is something unreal, existing outside the real world, in which everything
which is real is intentionally 'constituted'; it is something 'which has no
spatiotemporal exterior, and can be inside no spatiotemporal system; which cannot
experience causality from anything nor exert causality upon anything [ ... ] causality
as a relation of dependence between realities' .23
Point 1. belongs to epistemology, the rest are ontological or metaphysical
assertions. To pure consciousness leads, according to Hussserl, transcendental
phenomenological reduction which is supposed by hirn to open the proper sphere of
phenomenological inquiry, that of 'first philosophy'.
The results of Ingardens existential - and formal-ontological investigations in
Streit, together with some of his deliberations on the material (qualitative) aspects of
being, analyses anticipating his planned, but not executed systematic elaboration of
material ontology, have considerably changed the initial, based on Husserlian
concepts, perspective of his inquiry. Namely,
AI his analyses of temporal existence and his acceptance of the indubitable
temporality of conscious experiences, in particular their 'fissuration' , seem to make
impossible an absolute existence of human consciousness in the sense of its

'independence from any other thing'.
BI Ingarden's inquiry into the ontic structure of objects, in particular of whole
domains of objects, has shown that pure consciousness could be at most an object,
and not a domain of objects which it would have to be if transcendental idealism
were right.
CI A careful examination of the ontic situation of pure consciousness and of the
'pure ego' which, according to Husserl and Ingarden, is inseparable from it as the
'executor' of conscious acts shows that consciousness is 'inseparate,24 from the real
human person or, at the very least, is functionally united with the real subject. It also
turned out that 'pure' experiences must stand in a elose relation to the human body.
In this way Ingarden had to deny the possibility of the existence of pure
consciousness outside of the real world as something existing differently - and
'stronger' - than the world and, in particular, than a human being. If consciousness
existed differently than the world, its existential status would rather, according to
hirn, be lower than that of areal subject because of its dependence on it.
From among the four main characteristics which, according to Husserl, define
pure consciousness, it is only the first, epistemological one - namely the
indubitability of the existence and, at the very least, of some of the essential features
of pure consciousness which can be given in immanent perception - that has not
been undermined by Ingarden. His acceptance of this concept is substantiated solely
by epistemological reasons - by the postulate of indubitability of epistemological
assertions.
In the quoted par. 79 of Streit he writes:
[In the light of the above,l a certain line of thought has been opened which may allow us
- with the aid of material-ontological investigations - to move an essential step cJoser
to the solution [of the problem of the existence of the worldl. Namely, would it be
possible to ascertain in material-ontological investigation that pure consciousness is in
its essence existentially inseparate from or, at the very least. contingent25 upon the real
self and, in particular. on the soul; and would it, on the other hand, be possible to show
that - as Husserl claims - it is indeed, or may be. immanently given and, therefore. its
existence has the privilege of indubitability. then it would be also essentially easier to
acknowledge the existence of the 'real self. But we are still a preuy long way off this
result' ,>6
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.
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
sich nämlich in material-ontologischen Betrachtungen zeigen, dass das reine Bewusstsein

wesensmässig auf das reale Ich, und insbesondere auf die Seele unselbständig oder mindestens von ihr
abhängig ist. und ließe sich anderseits zeigen, dass das reine Bewusstsein - wie es Husserl behauptet
- wirklich immanent gegeben ist bzw. sein kann und sich somit einer ausgezeichneten
Seinsunbezweifelbarkeit erfreut, dann würde auch die Entscheidung bezüglich der Existenz des "realen
Ichs" wesentlich erleichtert werden. Aber davon sind wir noch ein Stück Weges entfernt'.
21 It is true !hat Ingarden did postulate a strict separation of epistemological problems from problems of
ontology and of metaphysics, and he certainly regarded his analyses of 'pure experiences' as genuine
descriptions and not as theoretical constructions. But his deep conviction about the correctness of the
'dogmatic motive' of Husserl's epistemology seems to have prevented hirn from drawing ultimate
conc\usions from his criticism ofhis master's conception of 'pure consciousness'.
28 See Ernst Tugendhat, Der Wahrheitsbegriff bei Husserl und Heidegger, 1967, Berlin: Oe Gruyter,
p. 194ff; see also my Swiat, spostrzezenie, swiadomosc. Fenomenologiczna koncepcja swiadomosci a
rea/izm. [Time, Perception, Consciousness. The Phenomenological Conception of Consciousness and
Realisml, 1973, Warszawa: PWN, p. 248ff.
29 See my 'Roman Ingarden, ein Metaphysiker der Freiheit', Repons on Pi/osophy (Krak6w) 10:1986,
p. 43-56; see also my 'Painting and the Structure of Consciousness. Remarks on Roman Ingarden's
Theory of Painting' in Kunst und Ontologie. hrsg. von W. Galewicz and W. Str6:iewski, 1994,
Amsterdam-Atlanta: Rodopi p. 80-95.
SYNTIIESE LIBRARY
1. J. M. BocMnski, APrecis of Mathematical Logic. Translated from French and German by O.

Bird. 1959 ISBN 90-277-0073-7
2. P. Guiraud, Problemes et methodes de la statistique linguistique. 1959 ISBN 90-277-0025-7
3. H. Freudenthal (ed.), The Concept and the Role ofthe Model in Mathematics and Natural and
Social Sciences. 1961 ISBN 90-277-0017-6
4. E. W. Beth, Formal Methods. An Introduction to Symbolic Logic and to the Study of Effective
Operations in Arithmetic and Logic. 1962 ISBN 90-277-0069-9
5. B. H. Kazemier and D. Vuysje (eds.), Logic and Language. Studies dedicated to Professor
Rudolf Carnap on the Occasion of His 70th Birthday. 1962 ISBN 90-277-0019-2
6. M. W. Wartofsky (ed.), Proceedings ofthe Boston Colloquiumfor the Philosophy ofScience,
1961-1962. [Boston Studies in the Philosophy of Science, Vol. I] 1963 ISBN 90-277-0021-4
7. A. A. Zinov'ev, Philosophical Problems of Many-valued Logic. A revised edition, edited and
translated (from Russian) by G. Koog and D.D. Comey. 1963 ISBN 90-277-0091-5
8. G. Gurvitch, The Spectrum ofSocialTIme. Translated from French and edited by M. Korenbaum
and P. Bosserman. 1964 ISBN 90-277-0006-0
9. P. Lorenzen, Formal Logic. Translated from German by F.J. Crosson. 1965
ISBN 9O-277~0080-X
10. R. S. Cohen and M. W. Wartofsky (eds.), Proceedings ofthe Boston Colloquiumfor the Philo-
sophy of Science, 1962-1964. In Honor of Philipp Franlc. [Boston Studies in the Philosophy
of Science, Vol. II] 1965 ISBN 90-277-9004-0
11. E. W. Beth, Mathematical Thought. An Introduction to the Philosophy of Mathematics. 1965
ISBN 90-277-0070-2
12. E. W. Beth and J. Piaget, Mathematical Epistemology and Psychology. Translated from French
by W. Mays. 1966 ISBN 90-277-0071-0
13. G. Küng, Ontology and the Logistic Analysis of Language. An Enquiry into the Contemporary
Views on Universals. Revised ed., translated from German. 1967 ISBN 90-277-0028-1
14. R. S. Cohen and M. W. Wartofsky (eds.), Proceedings of the Boston Colloquium for the
Philosophy of Sciences, 1964-1966. In Memory of Norwood Russell Hanson. [Boston Studies
in the Philosophy of Science, Vol. I1I] 1967 ISBN 90-277-0013-3
15. C. D. Broad, Induction, Probability, and Causation. Selected Papers. 1968
ISBN 90-277-0012-5
16. G. Patzig, Aristotle's Theory ofthe Syllogism. A Logical-philosophical Study of BookA ofthe
Prior Analytics. Translated from German by J. Barnes. 1968 ISBN 90-277-0030-3
17. N. Rescher, Topics in Philosophical Logic. 1968 ISBN 90-277-0084-2
Philosophy of Science, 1966-1968, Part I. [Boston Studies in the Philosophy of Science,
Vol. IV] 1969 ISBN 90-277-0014-1
Philosophy of Science, 1966-1968, Part 11. [Boston Studies in the Philosophy of Science,
Vol. V] 1969 ISBN 90-277-OO15-X
20. J. W. Davis, D. J. Hockney and W. K. Wilson (eds.), Philosophical Logic. 1969
ISBN 90-277-0075-3
21. D. Davidson andJ. Hintikka (eds.), Wontr and Objections. Essays on the Work ofW. V. Quine.
1969, rev. ed. 1975 ISBN 90-277-0074-5; Pb 90-277-0602-6
22. P. Suppes, Studies in the Methodology and Foundations ofScience. Selected Papersfrom 1951
to 1969. 1969 ISBN 90-277-0020-6
23. J. Hintikka, Models for Modalities. Selected Essays. 1969
ISBN 90-277-0078-8; Pb 90-277-0598-4
SYNTHESE LIBRARY
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
49. K. J. J. Hintikka, J. M. E. Moravcsik and P. Suppes (eds.), Approaches to Natural Language.

1973 ISBN 90-277-0220-9; Pb 90-277-0233-0
50. M. Bunge (ed.), Exact Philosophy. Problems, Tools and Goals. 1973 ISBN 90-277-0251-9
51. R. J. Bogdan and I. Niiniluoto (eds.), Logic, Language andProbability. 1973
ISBN 90-277-0312-4
52. G. Pearce and P. Maynard (eds.), Conceptual Change. 1973
ISBN 90-277-0287-X; Pb 90-277-0339-6
53. I. Niiniluoto and R. Thomela, Theoretical Concepts and Hypothetico-inductive lriference. 1973
ISBN 90-277-0343-4
54. R. Fraisse, Course 0/ Mathematical Logic - Volume 1: Relation and Logical Formula. Trans-
lated from French. 1973 ISBN 90-277-0268-3; Pb 90-277-0403-1
(For Volurne 2 see under No. 69).
55. A. Grünbaum, Philosophical Problems 0/ Space and Time. Edited by R.S. Cohen and M.W.
Wartofsky. 2nd enlarged ed. [Boston Studies in the Philosophy of Science, Vol. XII] 1973
ISBN 90-277-0357-4; Pb 90-277-0358-2
56. P. Suppes (ed.), Space, Time and Geometry. 1973 ISBN 90-277-0386-8; Pb 90-277-0442-2
57. H. Kelsen, Essays in Legal and Moral Philosophy. Selected and introduced by O. Weinberger.
Translated from German by P. Heath. 1973 ISBN 90-277-0388-4
58. R. J. Seeger and R. S. Cohen (eds.), Philosophical Foundations 0/ Science. [Boston Studies in
the Philosophy of Science, Vol. XI] 1974 ISBN 90-277-0390-6; Pb 90-277-0376-0
59. R. S. Cohen and M. W. Wartofsky (eds.), Logical and Epistemological Studies in Contemporary
Physics. [Boston Studies in the Philosophy of Science, Vol. :XIII] 1973
ISBN 90-277-0391-4; Pb 90-277-0377-9
60. R. S. Cohen and M. W. Wartofsky (eds.), Methodological and Historical Essays in the Natural
and Social Sciences. Proceedings 0/ the Boston Colloquium /or the Philosophy 0/ Science,
1969-1972. [Boston Studies in the Philosophy of Science, Vol. XIV] 1974
ISBN 90-277-0392-2; Pb 90-277-0378-7
61. R. S. Cohen, J. J. Stachel and M. W. Wartofsky (eds.), For Dirk Struik. Scientijic, Historical
and Political Essays. [Boston Studies in the Philosophy of Science, Vol. XV] 1974
ISBN 90-277-0393-0; Pb 90-277-0379-5
62. K. Ajdukiewicz, Pragmatic Logic. Translated from Polish by o. Wojtasiewicz. 1974
ISBN 90-277-0326-4
63. S. Stenlund (ed.), Logical Theory and Semantic Analysis. Essays dedicated to Stig Kanger on
His 50th Birthday. 1974 ISBN 90-277-0438-4
64. K. F. Schaffner and R. S. Cohen (eds.), PSA 1972. Proceedings o/the Third Biennial Meeting 0/
the Philosophy 0/ Science Association. [Boston Studies in the Philosophy of Science, Vol. XX]
1974 ISBN 90-277-0408-2; Pb 90-277-0409-0
65. H. E. Kyburg, Jr., The Logical Foundations 0/ Statistical In/erence. 1974
ISBN 90-277-0330-2; Pb 90-277-0430-9
66. M. Grene, The Understanding o/Nature. Essays in the Philosophy ofBiology. [Boston Studies
in the Philosophy of Science, Vol. XXIII] 1974 ISBN 90-277-0462-7; Pb 90-277-0463-5
67. J. M. Broekman, Structuralism: Moscow, Prague, Paris. Translated from German. 1974
ISBN 90-277-0478-3
68. N. Geschwind, Selected Papers on Language and the Brain. [Boston Studies in the Philosophy
of Science, Vol. XVI] 1974 ISBN 90-277-0262-4; Pb 90-277-0263-2
69. R. Fraisse, Course 0/ Mathematical Logic - Volume 2: Model Theory. Translated from French.
1974 ISBN 90-277-0269-1; Pb 90-277-0510-0
(For Volume 1 see under No. 54)
SYNTHESE LIBRARY
70. A. Grzegorczyk,An Outline olMathematical Logic. Fundamental Results and Notions explained
with all Details. Translated from Polish. 1974 ISBN 90-277-0359-0; Pb 90-277-0447-3
71. F. von Kutschera, Philosophy olLanguage. 1975 ISBN 90-277-0591-7
72. J. Manninen and R. Tuomela (eds.), Essays on Explanation and Understanding. Studies in the
Foundations ofHumanities and Social Sciences. 1976 ISBN 90-277-0592-5
73. J. Hintikka (ed.), Rudolf Camap, Logical Empiricist. Materials and Perspectives. 1975
ISBN 90-277-0583-6
74. M. Capek (ed.), The Concepts 01 Space and Time. Their Structure and Their Development.
[Boston Studies in the Philosophy of Science, Vol. XXII] 1976
ISBN 90-277-0355-8; Pb 90-277-0375-2
75. J. Hintikka and U. Remes, The Method 01 Analysis. Its Geometrical Origin and Its General
Significance. [Boston Studies in the Philosophy of Science, Vol. XXV] 1974
ISBN 90-277-0532-1; Pb 90-277-0543-7
76. J. E. Murdoch and E. D. Sylla (eds.), The Cultural Context 01 Medieval Leaming. [Boston
Studies in the Philosophy of Science, Vol. XXVI] 1975
ISBN 90-277-0560-7; Pb 90-277-0587-9
77. S. Amsterdamski, Between Experience and Metaphysics. Philosophical Problems ofthe Evol-
ution of Science. [Boston Studies in the Philosophy of Science, Vol. XXXV] 1975
ISBN 90-277-0568-2; Pb 90-277-0580-1
78. P. Suppes (ed.), Logic and Probability in Quantum Mechanics. 1976
ISBN 90-277-0570-4; Pb 90-277-1200-X
79. H. von Heimholtz: Epistemological Writings. The Paul Hertz / Moritz Schlick Centenary
Edition 011921 with Notes and Commentary by the Editors. Newly translated from German
by M. F. Lowe. Edited, with an Introduction and Bibliography, by R. S. Cohen and Y. Elkana.
[Boston Studies in the Philosophy of Science, Vol. XXXVII] 1975
ISBN 9O-277-0290-X; Pb 90-277-0582-8
80. J. Agassi, Science in Flux. [Boston Studies in the Philosophy of Science, Vol. XXVIII] 1975
ISBN 90-277-0584-4; Pb 90-277-0612-2
81. S. G. Harding (ed.), Can Theories Be Refuted? Essays on the Duhem-Quine Thesis. 1976
ISBN 90-277-0629-8; Pb 90-277-0630-1
82. S. Nowak, Methodology olSociological Research. General Problems. 1977
ISBN 90-277-0486-4
83. J. Piaget, J.-B. Grize, A. Szeminsska and V. Bang, Epistemology and Psychology 01 Functions.
Translated from French. 1977 ISBN 90-277-0804-5
84. M. Grene and E. Mendelsohn (eds.), Topics in the Philosophy 01 Biology. [Boston Studies in
the Philosophy of Science, Vol. XXVII] 1976 ISBN 90-277-0595-X; Pb 90-277-0596-8
85. E. Fischbein, The Intuitive Sources 01 Probabilistic Thinking in Children. 1975
ISBN 90-277-0626-3; Pb 90-277-1190-9
86. E. W. Adams, The Logic 01 Conditionals. An Application of Probability to Deductive Logic.
1975 ISBN 90-277-0631-X
87. M. Przel~ki and R. W6jcicki (eds.), Twenty-Five Years 01 Logical Methodology in Poland.
Translated from Polish. 1976 ISBN 90-277-0601-8
88. J. Topolski, The Methodology 0/ History. Translated from Polish by O. Wojtasiewicz. 1976
ISBN 90-277-0550-X
89. A. Kasher (ed.), Language in Focus: Foundations, Methods and Systems. Essays dedicated to
Yehoshua Bar-Hillel. [Boston Studies in the Philosophy of Science, Vol. XLIII] 1976
ISBN 90-277-0644-1; Pb 90-277-0645-X
SYNTIIESE 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
189. H. Siegel, Relativism Refuted. A Critique of Contemporary Epistemological Relativism. 1987

ISBN 90-277-2469-5
190. W. Callebaut and R Pinxten, Evolutionary Epistemology. A Multiparadigm Prograrn, with a
Complete Evolutionary Epistemology Bibliograph. 1987 ISBN 90-277-2582-9
191. J. Kmita, Problems in Historical Epistemology. 1988 ISBN 90-277-2199-8
192. J. H. Fetzer (ed.), Probability and Causality. Essays in Honor ofWesley C. Salmon, with an
Annotated Bibliography. 1988 ISBN 90-277-2607-8; Pb 1-5560-8052-2
193. A. Donovan, L. Laudan and R Laudan (eds.), Scrutinizing Science. Empirical Studies of
Scientific Change. 1988 ISBN 90-277-2608-6
194. H.R Otto and J.A. Thedio (eds.), Perspectives on Mind. 1988 ISBN 9O-277-2640-X
195. D. Batens and J.P. van Bendegem (eds.), Theory and Experiment. Recent Insights and New
Perspectives on Their Relation. 1988 ISBN 90-277-2645-0
196. J. Österberg, Self and Others. A Study of Ethical Egoism. 1988 ISBN 90-277-2648-5
197. D.H. Helman (ed.), Analogical Reasoning. Perspectives of Artificial Intelligence, Cognitive
Science, and Philosophy. 1988 ISBN 90-277-2711-2
198. J. Wolenski, Logic and Philosophy in the Lvov-Warsaw School. 1989 ISBN 90-277-2749-X
199. R W6jcicki, Theory 0/ Logical Calculi. Basic Theory of Consequence Operations. 1988
ISBN 90-277-2785-6
200. J. Hintikka and M.B. Hintikka, The Logic 0/ Epistemology and the Epistemology 0/ Logic.
Selected Essays. 1989 ISBN 0-7923-0040-8; Pb 0-7923-0041-6
201. E. Agazzi (ed.), Probability in the Sciences. 1988 ISBN 90-277-2808-9
202. M. Meyer (ed.), From Metaphysics to Rhetoric. 1989 ISBN 90-277-2814-3
203. RL. Tieszen, Mathematical Intuition. Phenomenology and Mathematical Knowledge. 1989
ISBN 0-7923-0131-5
204. A. Melnick, Space, Time, and Thought in Kant. 1989 ISBN 0-7923-0135-8
205. D.W. Smith, The Circle 0/ Acquaintance. Perception, Consciousness, and Empathy. 1989
ISBN 0-7923-0252-4
206. M.H. Salmon (ed.), The Philosophy 0/ Logical Mechanism. Essays in Honor of Arthur W.
Burks. With his Responses, and with a Bibliography of Burk's Work. 1990
ISBN 0-7923-0325-3
207. M. Kusch, Language as Calculus vs. Language as Universal Medium. A Study in Husserl,
Heidegger, and Gadamer. 1989 ISBN 0-7923-0333-4
208. T.C. Meyering, Historical Roots 0/ Cognitive Science. The Rise of a Cognitive Theory of
Perception from Antiquity to the Nineteenth Century. 1989 ISBN 0-7923-0349-0
209. P. Kosso, Observability and Observation in Physical Science. 1989 ISBN 0-7923-0389-X
210. J. Kmita, Essays on the Theory o/Scientific Cognition. 1990 ISBN 0-7923-0441-1
211. W. Sieg (ed.), Acting and Rejlecting. The Interdisciplinary Thrn in Philosophy. 1990
ISBN 0-7923-0512-4
212. J. KarpinSki, Causality in Sociological Research. 1990 ISBN 0-7923-0546-9
213. H.A. Lewis (ed.), Peter Geach: Philosophical Encounters. 1991 ISBN 0-7923-0823-9
214. M. Ter Hark, Beyond the Inner and the Outer. Wittgenstein's Philosophy of Psychology. 1990
ISBN 0-7923-0850-6
215. M. Gosselin, Nominalism and Contemporary Nominalism. Ontological and Epistemological
Implications of the Work of W. V.O. Quine and of N. Goodman. 1990 ISBN 0-7923-0904-9
216. J.H. Fetzer, D. Shatz and G. Schlesinger (eds.), Definitions and Definability. Philosophical
Perspectives. 1991 ISBN 0-7923-1046-2
217. E. Agazzi and A. Cordero (eds.), Philosophy and the Origin and Evolution 0/ the Universe.
1991 ISBN 0-7923-1322-4
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
290. G. Sommaruga: History and Philosophy of Constructive Type Theory. 2000

ISBN 0-7923-6180-6
291. J. Gasser (ed.): ABooie Anthology. Recent and Classical Studies in the Logic of George Boole.
2000 ISBN 0-7923-6380-9
292. V.F. Hendricks, S.A Pedersen and K.F. Jl'lrgensen (eds.): Proof Theory. History and Philo-
sophical Significance. 2000 ISBN 0-7923-6544-5
293. W.L. Craig: The Tensed Theory ofTime. A Critical Examination. 2000 ISBN 0-7923-6634-4
294. W.L. Craig: The Tenseless Theory ofTime. A Critical Examination. 2000
ISBN 0-7923-6635-2
295. L. Albertazzi (ed;): The Dawn ofCognitive Science. Early European Contributors. 2001
ISBN 0-7923-6799-5
296. G. Forrai: Reference, Truth and Conceptual Schemes. A Defense oflntemal Realism. 2001
ISBN 0-7923-6885-1
297. Y.F. Hendricks, S.A Pedersen and K.F. Jl'lrgensen (eds.): Probability Theory. Philosophy,
Recent History and Relations to Science. 2001 ISBN 0-7923-6952-1
298. M. Esfeld: Holism in Philosophy of Mind and Philosophy of Physics. 2001
ISBN 0-7923-7003-1
299. E.C. Steinhart: The Logic of Metaphor. Analogous Parts of Possible Worlds. 2001
ISBN 0-7923-7004-X
300. To be published.
30l. T.AF. Kuipers: Structures in Science Heuristic Patterns Based on Cognitive Structures. An
Advanced Textbook in Neo-Classical Philosophy of Science. 2001 ISBN 0-7923-7117-8
302. G. Hon and S.S. Rakover (eds.): Explanation. Theoretical Approaches and Applications. 2001
ISBN 1-4020-0017-0
303. G. Holmström-Hintikka, S. Lindström and R. Sliwinski (eds.): Collected Papers ofStig Kanger
with Essays on his Life and Work. Vol. I. 2001
ISBN 1-4020-0021-9; Pb ISBN 1-4020-0022-7
304. G. Holmström-Hintikka, S. Lindström and R. Sliwinski (eds.): Collected Papers ofStig Kanger
with Essays on his Life and Work. Vol. 11. 2001
ISBN 1-4020-0111-8; Pb ISBN 1-4020-0112-6
305. C.A Anderson and M. Zeleny (eds.): Logic, Meaning and Computation. Essays in Memory
of Alonzo Church. 2001 ISBN 1-4020-0141-X
306. P. Schuster, U. Berger and H. Osswald (eds.): Reuniting the Antipodes - Constructive and
Nonstandard Views ofthe Continuum. 2001 ISBN 1-4020-0152-5
307. S.D. Zwart: Refined Verisimilitude. 2001 ISBN 1-4020-0268-8
308. A-S. Maurin: IF Tropes. 2002 ISBN 1-4020-0656-X
309. H. Eilstein (ed.): A Collection of Polish Works on Philosophical Problems ofTime and Space-
time. 2002 ISBN 1-4020-0670-5
310. Y. Gauthier: Internal Logic. Foundations ofMathematics from Kronecker to Hilbert. 2002
ISBN 1-4020-0689-6
311. E. Ruttkamp: A Model- Theoretic Realist Interpretation of Science. 2002
ISBN 1-4020-0729-9
Previous volumes are still available.
KLUWER ACADEMIC PUBLISHERS - DORDRECHT I BOSTON I LONDON

A Collection of Polish Works On Philosophical Problems of Time and Spacetime

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

A Collection of Polish Works On Philosophical Problems of Time and Spacetime

Hochgeladen von

Copyright:

Verfügbare Formate

A COLLECTION OF POLISH WORKS ON PHILOSOPHICAL PROBLEMS

OF TIME AND SPACETIME

LOGIC, METHODOLOGY, AND PHILOSOPHY OF SCIENCE

JAAKKO HINTIKKA, Boston University, U.S.A.

DIRK VAN DALEN, University of Utrecht, The Netherlands

Springer-Science+Business Media, B.Y.

ISBN 978-90-481-6039-6 ISBN 978-94-017-0097-9 (eBook)

Printed on acid-free paper

All Rights Reserved

This is a collection of some works of Polish philosophers and physicists on

MOTION, SPACE, TIME*.

H. Eilstein (ed.). A Collection 0/ Polish Wor/es on Philosophical Problems

According to relationists the adequate theory of motion should contain in its

Spacetime properties, i.e. spacetime symmetries acknowledged by a given

2. THE PRERELATIVISTIC PHYSICS

where R aß is a constant orthogonal matrix, va = const and a, ß =1,2,3. Notation

(where m - mass of a particle, Pz - an impressed force, x a - location of the

where M -' differential manifold, Ai - absolute objects and Pi - dynamical

The introduction in Newton's Scholium, of a distinguished reference frame into

t ~ t' =/(t) , dfldt >0

Leibnizian symmetries, in turn, have the form:

where F= (4<jmjmj rj/ 2/12, 'P= 4<jmj mjlrjj,

mjd x;' I dt = ( 11 M'P) . fJ'Pl fJXj (4)

3. THE RELATIVISTIC PHYSICS

principle of relativity, tells that the laws of physics (the equations

where Xl =ct, ~ =X, ~ =y, x4 =z.

LiT = .Jc LJ!,2 =cL1t'

T(t) = je~1-v2 fe 2dt' (9)

where v is the velocity of the particle with respect to S.

where / is the energy-momentumJour-veetor of the particle, mo the rest mass of this

above, according to the Noether's theorem, every spacetime symmetry corresponds

where Rj - Ricci tensor, R - Ricci curvature scalar, G - gravitation al constant,

This equation of motion becomes simplified in a local inertial (freely falling)

Since Mach's program remains unfulfilled, there is no possibility to connect the

4. ONTOLOGICAL IMPLICATIONS OF THE CONTROVERSY BETWEEN

-REL=:> SUB (16)

-SUB~ REL (17)

differentiable transformations. These symmetry transformations make it possible

QUANTUM SPACETIME AND THE PROBLEM

1. WHY TO QUANTIZE GRAVITY?

appear. More specifically: what is quantum spacetime, and in particular, what

2. TIME AND SPACE IN GENERAL RELATIVITY.

In GR time is not a fundamental entity, as it was according to Newton in non-

be highly distorted in comparison to the genuine cone. We focus our attention on

preferred in comparison to the others. For example: in SR by choosing an inertial

3. TIME AND QUANTUM THEORY. QUANTUM SPACETIME.

probability of any possible outcome of any experiment (measurement) performed on

not a theory of gravitational interactions taking place on a fixed spacetime

'foamlike' picture of spacetime is suggestive, it is now believed that it is too naive

4. QUANTUM THEORY ON A FIXED CURVED SPACETIME

energy (while the corresponding solutions with negative energy represent

5. QUANTUM COSMOLOGY AND FOUNDATIONS

The physicist's views on the conceptual foundations of quantum theory (particularly

that it predicts probabilities of various effects, e.g. a probability that in a collision of

equations. However this overabundance of models makes some physicists feel

Until the end of 201h century investigations of quantum cosmology, though

Acknowledgement. This work was partially supported by a Polish Committee

TIME AND PHYSICS - A NONCOMMUTATIVE

Modern philosophy of time is inseparably connected with the progress in modern

met in various attempts to construct quantum gravity or final unifying theories.

2. SOURCES OF NONCOMMUTATIVE GEOMETRY

time evolution of observables in quantum mechanics is also defined in terms of

6. TIME AND PROBABILITY