356 DOC.
364
JULY
1917
sky
does not rotate
is
taken
as a
basis,
can
a
system
in which
the Earth’s
center
is
at
rest
possibly
be chosen for
the
simplest description
of
the
phenomena.
All these considerations
on
the
selection of
a
system
in which
the
phenomena
can
be
represented
in
the
simplest
way, might possibly appear
to
the
physicist
as
entirely
inappropriate
to his
field,
especially
the
preference
of
a system
with
respect
to uniform
velocity. Physicists
can
merely
deal with
an
abstraction
of
the
real
processes,
which alone
can
be
generalized
(in
the
form
of
a
differential
equa-
tion)
and
which
requires
“arbitrary”
initial
conditions for its
integration.
Could
this
ever
be
otherwise,
or
is
this
founded
on
the
nature
of
the
world? Does
chance,
in
this
sense, incapable
of
being
subjected
to
laws,
really
exist in
the
universe? Or
will
it
perhaps
be
possible
one
day
to
discover
laws
about the distribution
of actu-
ally existing
forms in
the
four-dimensional
world,
such
as,
in
the
case
of
an
infinite
universe,
of
a periodic
nature
in
space
and
time?
In
my cosmological paper,
which
I
am sending you,[6]
the
sole
evidence
is
the
probability concept;
however,
it
is
not
[un]thinkable,
in
principle, although
altogether
improbable,
that
it would be
possible
to find
an
integrated
formula of
the
world,
without
arbitrary constants,
that
describes
the universe
as
a
whole and
allows
us
to
identify
the
patterns
in
its four-dimensional
structure. As
this
is
currently
not
the
case, my
reflections
on
the coordinate
system may
be
relegated
to
epistemology
or
to
cosmography
and
astronomy.
Nonetheless,
my
and
Adler’s idea
does
not
seem
totally
irrelevant,
because
it
specifically points
out
that the
special
or
general
covariance of
the
nat-
ural
laws
does
not
exclude that,
even
with
regard to
uniform
motion,
a
specific
system
is
preferred
for
the
description
of
the world of the fixed-star
system
known
to
us.
If
one
wants to
call this
simplest system
the
“true”
one,
no
objections
can
be
raised,
even
from
the
standpoint
of
the
physical
theory
of
relativity,
but
only
from
an
epistemological
and
cosmological
point
of
view.
However,
the
possibility
of
denoting
a
system
as
ultimately
“true”
only
remains if
the material
cosmos
is
finite,
since
otherwise,
in
considering increasingly comprehensive
worlds,
we
are
going
to
be forced
to
change
over
to
other
preferred coordinates,
after
all.
As
I
myself
do not believe in
the
boundedness of the
world, my
ideas do
not
in
any
way
guide me
toward
a
contradiction
to
the
epistemological principle
of
relativity.
You
probably
gather
from all
of
the
above
that
I
furthermore take
a
purely phenomenalistic
stand
and therefore do not
believe
that
any
coordinate
system
is
the
true
one,
in
a metaphysical sense.
But
the
epistemological princi-
ple
of
relativity
does
not
necessarily
follow
from
the
validity
of all
your physical
theories.
Essentially,
the
physical
theory
of
relativity actually
merely
involves
(1)
theoretically
deriving
new
and unforeseen
facts,
determined
by
natural
laws,
from
certain
general assumptions on
the
representation
of
the
phenomena
from
various
points
of view
(systems);
(2)
providing
us
with
the
means
to
calculate
directly
the
phenomena
for
an
observer
moving
with
reference to
one
observer