Black holes: a prediction of theory or phantasy?

A.A. Logunov*, M.A. Mestverishvili*, V.V. Kiselev*

* Russian State Research Center “Institute for High Energy Physics”, Pobeda 1,
arXiv:gr-qc/0412058 v1 13 Dec 2004

Protvino, Moscow Region, 142281, Russia

E-mail: kiselev@th1.ihep.su

Abstract. We argue for black holes do not represent a strict consequence of general
relativity.

In Einstein general relativity a gravity is described by a metric tensor gµν of pseudo-

Riemannian space-time. An interval of space-time has a generic form
ds2 = gµν dxµ dxν , with the signature (−2). (1)
As usual, let us isolate time-like and space-like parts in (1)
ds2 = c2 dτ 2 − dℓ2 , (2)
with
1 g0ν dxν
dτ = √ (3)
c g00
representing a physical time, while the expression
g0i g0k
dℓ2 = κik dxi dxk , κik = − gik (4)
g00
gives a square of space distance between two infinitely close points. The metric
coefficients gµν have to satisfy well-known inequalities by Hilbert [1, 2].
A.Einstein especially noted [3]: “Therefore, there is no way except to recognize
all thinkable‡ coordinate systems to be, in principle, equivalent for the description of
nature”. We will further follow this statement by A.Einstein.
A motion of probe bodies takes place on time-like geodesics. In the pseudo-
Riemannian space-time, a time-like geodesic curve, by definition, cannot become
isotropic at a point. From the basic statement of theory on the pseudo-Riemannian
geometry of space-time one unavoidably deduces that a physical solution of gravitation
equations at any time-like geodesic curve should satisfy the following inequality:
 dℓ 2
v2 = < 1. (5)
dτ
It means that the physical velocity cannot be equal to the speed of light at any point of
trajectory (including those at infinity). The equality v 2 = 1 implies the interval becomes
‡ We will not concern for some constraints caused by requirements of unambiguity and continuity.
(A.Einstein)
Black holes: a prediction of theory or phantasy? 2

isotropic, i.e. it equals zero. For a physical solution of gravitation equations the
time-like interval does not permit the physical velocity equal to the speed of light. The
inequality of (5) should be valid in any system of coordinates. If it is broken in a system
of coordinates (or reference frame) for a time-like geodesic curve even in a single point,
then the whole solution contradicts to primary principles of the theory. The solution
has the physical sense in the validity region of (5), only.
Let us consider the Schwarzschild solution. K. Schwarzschild found the following
exact spherically symmetric solution of gravitation equation in January, 1916:
r − α  r 
ds2 = dt2 − dr 2 − r 2 (dθ2 + sin2 θdφ2 ) , (6)
r r−α
where
2GM
α= . (7)
c2
In addition to the singularity at r = 0 the solution has the singularity at r = α. This
singularity was called “the Schwarzschild singularity” or “Hadamard catastrophe”.
Having considered a radial motion of probe particle on a time-like geodesic curve
in the Schwarzschild metric, in 1917 D. Hilbert has obtained the equation [1]
d2 r 3α  dr 2 α(r − α)
− + = 0, (8)
dt2 2r(r − α) dt 2r 3
and found the integral of motion A,
 dr 2  r − α 2  r − α 3
= +A , (9)
dt r r
where A is a constant determined by initial data. By (4) and (6) the physical velocity
of particle is equal to
r 2  dr 2
v2 = . (10)
(r − α)2 dt
Substituting it to equation (9), we get
r−α
v2 = 1 + A . (11)
r
For the time-like geodesic, the constant A can take values in the region
r0
− ≤ A < 0, (12)
r0 − α
where r0 > α or r0 < 0, and A = 0 corresponds to the light.
From (11) we can see that on the radial time-like geodesics at r = α the physical
velocity equals the speed of light. Thus, the Schwarzschild solution at r = α breaks
inequality (5), converting the time-like curve into the isotropic one. However, this
fact is not consistent with the general relativity. Therefore such the solution is valid at
r > α, only.
The analogous fact can be observed, if by the transformation
√  √r − √ α 
t = τ + 2 αr + α ln √ √  (13)
 
r+ α
Black holes: a prediction of theory or phantasy? 3

we arrive to the interval in the form of Painlevé [4]

r − α r
2 2 α
ds = dτ + 2 drdτ − dr 2 − r 2 (dθ2 + sin2 θdφ2 ) . (14)
r r
Then, the physical velocity equals
2 dr
 2
r
v2 = h dτ , (15)
√ dr i2
(r − α) + αr
dτ
so that, for instance, at A = −1 we get on the radial geodesic curve
r r 
dr α dr α r − α
= , =− . (16)
dτ r dτ r r+α
The physical velocity of (15) becomes equal to
α
v2 = . (17)
r
It is again equal to the speed of light at r = α, hence, the solution in this form contradicts
to the primary fundamental statement of the theory in the form of (5).
In discussion devoted to the Schwarzschild singularity in 1939 A. Einstein wrote
in [4]: “The basic result of study is the distinct understanding that there are no
”Schwarzschild singularities” in the real world. Thought the offered theory considers
only such systems, in which particles move by circle trajectories, we have to hardly doubt
that the investigation of generic case will lead to the same results. The Schwarzschild
singularity is absent, because the matter cannot concentrated in an arbitrary way;
otherwise particles composing the clump will reach the speed of light”.
We have been convinced for this fact by having got eqs. (11) and (17) for the
physical velocity. By the opinion of A. Einstein, such the singularity is not admissible
in the vacuum, since it contradicts the general relativity. However, if it does appear,
then the solution itself has a restricted or constrained sense. It is valid at r > α, only.
The above caution by A. Einstein remained with no due attention and analysis.
A further analysis of Schwarzschild solution went in the way of searching for such
coordinates, in which the metric coefficients would be regular at r = α. Making use of
singular transformations the corresponding procedure was performed. So, for example,
transforming to new coordinates τ, R by formulae of [2, 6]
Z √ Z √
rdr rr
τ =t+ , R =t+ dr , (18)
r−1 r−1
(here we put units of 2GM/c2 = 1) interval (6) takes the form
1
ds2 = dτ 2 − (dR)2 − r 2 (dθ2 + sin2 θdφ2 ) , (19)
r
where
h3 i2/3
r= (R − τ ) . (20)
2
In coordinates τ, R, the singularity of metric coefficients is absent at r = 1. The
singularity is present at r = 0.
Black holes: a prediction of theory or phantasy? 4

For interval (19) the physical velocity is calculated by the formula

2 1  dR 2
v = . (21)
r dτ
On the basis of (9), let us choose the radial motion at A = −1, when the velocity at the
infinity is equal to zero:
dr 1 r − 1
=√ . (22)
dt r r
By (18) and (22), we get
dt r
= .
dτ r+1
Using the above equality and (22), we find
dr 1 r − 1
=√ . (23)
dτ r r+1
Then, by (20) we derive
dR 2r
= . (24)
dτ r+1
Substituting (24) into (21), for interval (19) we get the physical velocity
4r
v2 = . (25)
(r + 1)2
Therefore, we see that in terms of τ, R the physical velocity at r = 1 is equal to the
speed of light, too, i.e. the time-like geodesic curve becomes isotropic at this point.
Then, by transformation (18) the Schwarzschild singularity at r = 1 disappears in the
geodesic radial motion
dr 1 r − 1
= −√ ,
dt r r
but it still remains on the geodesic curve
dr 1 r − 1
=√ .
dt r r
It is absolutely evident that by singular transformations it could be removed from r = 1,
nevertheless, it will unavoidably appear at another point.
Although in this case all of metric coefficients in (19) are regular outside the source,
nevertheless, inequality (5) is broken on some geodesic curves. This fact witnesses that
the Schwarzschild solution has a constrained sense in the general relativity, therefore, its
continuation to other regions has a formal mathematical character. Then, black holes
invented about fifty years ago in scientific articles represent a product of pure phantasy,
since they have no solid theoretical ground.
Now consider singular transformation by Kruskal. The Kruskal variables [7] are the
followings:
√ t
U = r − 1 er/2 cosh ,
2 (26)
√ t
V = r − 1 er/2 sinh .
2
Black holes: a prediction of theory or phantasy? 5

In the case of radial motion, interval (6) takes the form

4
ds2 = e−r (dV 2 − dU 2 ) . (27)
r
In the Schwarzschild interval the component g00 is equal to
r−1
g00 = ,
r
which gives its nullification at r = 1, while in the Kruskal coordinates of (27) the
component g00 equals
4
g00 = e−r
r
and it tends to zero at r → ∞. For the radial motion on the time-like geodesic curve
at A = −1
dr 1 r − 1
=√ . (28)
dt r r
In terms of U and V the physical velocity equals
√
dU V r+U
= √ . (29)
dV U r+V
In accordance with (28) after the integration we get, for instance,
√
r−1 2 √
t = β(r) + ln √ , β(r) = r 3/2 + 2 r . (30)
r+1 3
By (26) and (30) we deduce
 √r − 1  U + V
t β
e =e √ = , (31)
r+1 U −V
wherefrom we obtain
√ √
V eβ ( r − 1) − ( r + 1)
= β √ √ . (32)
U e ( r − 1) + ( r + 1)
Substituting (32) into (29), we find
dU eβ(r) − 1
= β(r) . (33)
dV e +1
For the other radial geodesic motion
dr 1 r − 1
= −√ . (34)
dt r r
In the same way we easily get the following expression for the physical velocity:
dU eβ(r) − 1
= − β(r) . (35)
dV e +1
From (33) and (35) we deduce that at r → ∞ the physical velocity on the geodesic
time-like curve becomes equal to the speed of light, and the time-like geodesic curve
becomes isotropic.
The Schwarzschild singularity of metric coefficients at the sphere r = 1 is formally
cancelled by the Kruskal transformation. Nevertheless, the solution of Hilbert–Einstein
Black holes: a prediction of theory or phantasy? 6

equations does not become physical, since in accordance to the solution the physical
velocity became equal to the speed of light at infinity, which is unacceptable. In
the Schwarzschild variables the physical velocity at infinity of r is always less than
the speed of light, while at r = 1 it is equal to the speed of light. In the Kruskal
coordinates the situation is slightly different: the physical velocity equals the speed of
light at infinity of r, while it is always less than the speed of light at r = 1. The
presence of Schwarzschild singularity is not consistent with the general relativity.
Thus, the exact spherically symmetric solution of gravitation equations is physically
inconsistent in both the Schwarzschild coordinates and Kruskal ones due to the
presence of Schwarzschild singularity, since the solution breaks constraint (5). But
the concept of “black holes” is based on such the solution. In other words, the
Schwarzschild singularity led to “black holes”. On another hand, the presence of such
the singularity contradicts to the general relativity. Therefore, it is not true to accept
“black holes” as a consequence of general relativity. The above analysis has shown that
the Schwarzschild singularity cannot be removed by coordinate transformations, since
it is related with the nullification of interval ds2 . The singularity can be removed from
the metric coefficients, but not from the interval. Thus, we see that the notion of “black
holes” is based on the presence of Schwarzschild singularity, which is in contradiction
with the basics of general relativity, i.e. the pseudo–Riemannian geometry of space-time.

Finally, authors express a sincere gratitude to S.S. Gershtein, V.A. Petrov, and
N.E. Tyurin for valuable discussions.

References

[1] Hilbert D., Nachr. Ges. Wiss. Göttingen, Math-Phys. Klasse. 1916. 23.
[2] Landau L.D., Lifshitz E.M., “Field Theory”, Moscow: Fizmatlit, 2001, p. 536.
[3] Einstein A., Collection of scientific papers, Moscow: Nauka, 1966, I, p. 459 (in Russian).
[4] Painlevé P., Comptes rendus, Académie des sciences. 1921, 173 (17), pp. 667–680.
See also Gravity. 2001, 5 (1), pp. 16–27 (in Russian).
[5] Einstein A., Collection of scientific papers, Moscow: Nauka, 1966, II, p. 878 (in Russian).
See also Einstein A., Ann. Math. 1939, 40, pp. 922–936.
[6] Rylov Yu.A., Zh.E.T.P., 1961, 40, pp. 1755–1757 (in Russian).
[7] Kruskal M.D., Phys. Rev. 1960, 119, 1743