The modern creationist
movement has made much progress over the past 30 years. This has coincided
with the introduction of several periodicals and many books by numerous
authors and various publishers. The general approach has been two-fold:
first, to show that many observed properties of the world cannot adequately
be explained by evolutionary or uniformitarian models, and two, to show
that things can be explained better by the sudden creation of the world
in the not so distant past. The second step has properly shied away
from the response "that is the way God did it," in favor of
the design and order that a benevolent Creator has ordained. When processes
have been invoked, they have been ones of steady degeneration or of
The goal of all of this
effort has been to produce a consistent and detailed alternate model
that is Biblically correct and adequately describes what we observe.
This work has progressed in the fields of biology and geology so that
today we have a fairly well defined creationist model of each. The reviews
of Gish (1975, 1989a, 1989b) show that these two fields have consumed
the most attention of Quarterly authors since the inception of
the Creation Research Society.
Astronomy is another science
in which evolutionary and uniformitarian assumptions have had great
sway, however a survey of creationist literature reveals that much less
work has been done in this field as compared to biology and geology.
Discussion has mainly involved five key topics:
Initial origin of the
universe and solar system members
Age of things
Life in space
Anthropic principle of design
Of these, most of the
work that has been done has fallen into two areas. One of these is the
examination of the Moon, planets, comets, and other solar system objects
to argue that they must have young ages (Humphreys, 1990). The other
area leaps to the grand scale of the universe (cosmology) to argue against
the standard, or Big Bang, model of the universe (DeYoung and Whitcomb,
1981). Since Big Bang cosmogony demands an ancient age for the universe,
the purpose of this assault has obviously been to demonstrate that the
universe does not necessarily have to be as old as currently thought.
We offer three criticisms
of the present state of creationist thought in astronomy. First, as
stated above, most attention has centered on the small scale (the solar
system) and the large scale (cosmology), while leaving a rather large
gap in scale between these. For example, the second Creation Research
Society Monograph (Mulfinger, 1983) on design and origins in astronomy
has only two sections of contributed papers, one on the universe and
one on the solar system. Second, there has not yet emerged even a rough
framework of an alternative creation model as we find in biology and
geology. Third, except for teleological ramifications of the Earth's
immediate environment, there does not seem to be any stress upon the
purpose to the order and structure that we see in the universe. This
last point seems to leave us with the uncomfortable position that the
universe is as it is because God simply wanted it that way. As stated
earlier, this has not been the general attitude prevalent in creationist
studies of either biology or geology.
Actually, all three criticisms
stem from the same root of a lack of a reasonably complete and consistent
model, and the key seems to be the middle scale of criticism two. In
general astronomical parlance this middle scale would be referred to
as "stellar evolution" which purports to explain the structure
of stars and to explain how and why they got to be in their present
states. The use of the word "evolution" here carries a different
connotation as it does in biology where an increase in complexity is
implied. Instead, its astronomical use means change, whether the change
is one of decay or perceived improvement. Furthermore, the evolution
or change of stars is quite quantitative and is based upon well understood
physical principles such as hydrostatic equilibrium, thermal equilibrium,
equations of state of gases, and nuclear reactions. Still, the concept
of stellar evolution is an attempt to explain the observed properties
of the universe apart from the input of a creator. This, of course,
should suggest caution to creationists. Indeed, Mulfinger (1973) has
. . . many professing
Christians are being 'carried along with the tide.' Surely they fail
to realize the consequences. There is no logical stopping point. The
theory is a broad philosophical 'pathway' leading ultimately to atheism.
Because of this danger,
some creationists maintain the absolute or nearly absolute fixity of
stars. The well observed occurrence of novae and supernovae should show
that some change or evolution occurs. It is usually responded that these
processes demonstrate that the evolution that we see is one of degeneration.
But does not that conclusion stem from the general belief that such
eruptions are from dying stars? And how do we know this? The study of
stellar evolution tells us. Many creationists seem quite willing to
accept certain conclusions of stellar evolution, while rejecting out
of hand much of the theory leading to these same conclusions.
We feel that the whole
topic of stellar evolution needs detailed examination from a creationist
viewpoint, and it is hoped that this article and others to come will
spark much interest and discussion in this journal and elsewhere. To
be quite fair, there have been some attempts in the past, but they have
not been of the scope that we feel is necessary. For instance, Wilt
(1983) in his discussion of nucleosynthesis briefly described the major
aspects of stellar evolution without offering many specific criticisms
or alternatives. Mulfinger (1970) in a critique of stellar evolution
identified a few problems in the theory current over 20 years ago; we
are recommending that this work be updated and expanded.
As stressed earlier, one
of the first issues that should be discussed is how much change in stars
should we be willing to grant. It is not prudent to come to a hasty
decision on this subject. Stellar evolution computer codes begin with
models of stellar interior structure, which in turn rely upon well understood
and quantitative laws of physics. Complete rejection of stellar evolution
would erode confidence in current understanding of stellar structure
and would seem to repudiate much of physics as well. If creationists
wish to scrap stellar evolution completely, then it is incumbent on
us to rework stellar structure and/or physics in a convincing fashion.
The present authors are not entirely certain about what, if any, should
be kept, and we welcome discussion with other interested parties on
of Stellar Structure and Evolution
A brief summary of stellar
structure and evolution would be appropriate at this point. For further
study the reader is directed to the following recent review by Iben
(1991) or to the standard texts in the field: Clayton (1968), Cox and
Giuli (1968), Kippenhahn and Weigert (1990), Novotny (1973) and Schwarzchild
(1958). We will not discuss here the birth and early development of
stars, but instead start during the stable part of a star's lifetime.
The fundamental problems with stellar birth have been discussed previously
and perhaps we will return to this issue in a future paper.
The standard observational
tool used in studying stellar structure and evolution is the Hertzsprung-Russell
(H-R) diagram, so called because it was independently discovered by
two astronomers by those names early in this century. It consists of
a plot of stellar luminosities versus stellar temperatures, with luminosity
increasing upward and temperatures increasing to the left. Both quantities
are plotted logarithmically. A schematic diagram is shown in Figure
Figure 1. Schematic
Hertzsprung-Russell Diagram. L is star luminosity; T is temperature.
Most stars are found on
a roughly diagonal band called the main sequence (MS), while some stars
are found in the upper right of the diagram and a few are found to the
lower left. The former are called red giants, while the latter are called
white dwarfs. These names are appropriate because of the colors of these
stars (reflecting temperature) and their sizes, which can be deduced
with the following equation:
L = R2T4,
where L is a star's luminosity,
R is the radius, and T is the surface temperature, and all variables
are in solar units (L = R = T = 1, for the sun). Notice that the above
equation is very strongly dependent upon the temperature: a star to
the right of the HR diagram has a low temperature and hence should have
a low luminosity, while a star to the left has a high temperature and
so would have a high luminosity. Stars to the upper right and lower
left violate this trend, and, and this can only be if they are very
large and very small respectively. White dwarfs are about the size of
the earth, about 100 times smaller than the sun, while red giants may
be hundreds of times larger than the sun.
A stellar interior model
gives quantitative values of several physical quantities, such as temperature,
density and pressure, at regularly spaced intervals inside a star. There
are several assumptions, such as spherical symmetry, hydrostatic equilibrium,
thermal equilibrium, and equation of state, nuclear energy generation
processes, and radiative and convective energy transport. Each of these
principles comes from well understood physics and finds applications
in other fields. The equations can be expressed in differential form
and must be solved simultaneously with boundary conditions that the
star has the observed luminosity and temperature at its surface. Because
the solution relies upon the use of a numerical integration scheme,
much progress has been made since the invention of modern computers.
Early in the development
of astrophysics the Russell-Vogt theorem was proved, which states that
the structure of a star and hence its location on the H-R diagram is
determined by the star's mass and composition. Further, it was shown
that MS stars are consistent with models of stars having general cosmic
abundances (mostly hydrogen and helium, with only a few percent by mass
of everything else) and deriving their energy from hydrogen to helium
conversion in their cores. Creationists have described possible gravitational
contraction of the sun and other stars (DeYoung and Rush, 1989). This
may indeed be occur in certain stars, but the extreme temperature and
pressure of stellar interiors assures that nuclear fusion is a major
energy source. Even with only simple models this can be shown, as well
as several possible predictions about the MS. First the most massive
stars are found to the upper left of the MS, while the masses gradually
decrease toward the lower right. Furthermore, there should be a relationship
between the mass and the radius (M-R relation) and a relationship between
the mass and luminosity (M-L relation). Stellar masses can be determined
from studying binary stars, while radii and luminosities can be deduced
from a couple of different method each. Good observational data exist
for a few score MS stars, to which the three model predictions agree
This agreement is quite
impressive and the physical assumptions that go into it are so well
founded that it is doubtful that many creationists would have much to
argue with in MS structure. However, what is generally called post MS
evolution is not far removed from the brief outline of stellar structure
given above. As thermonuclear reactions occur, the composition of a
star will gradually change over time. A grid of stellar interior models
for a given mass but reflecting the composition changes shows the gradual
development or evolution that one would expect from the models. It would
appear that acceptance of stellar structure involves a step or two down
the slippery slope to which Mulfinger's quote referred. With that warning,
let us turn to post MS development as suggested by stellar structure.
We are not denying that stars were made on the fourth day of creation,
complete with variety and maturity. This view of an instantaneous, fully
functioning universe can be readily be built into a creationist model
of astronomy. The point that stars, in whatever initial stage of development,
will naturally change due to energy considerations, if time permits.
The rate of change, of course, is a critical variable of great interest
Recall that the Russell-Vogt
Theorem states that a star's mass and composition determine where a
star is located on the H-R diagram. On the MS the compositions of stars
are believed to be about the same (mostly hydrogen and very little of
elements heavier than helium), so that mass is the only important determinant.
Eventually all of the hydrogen in a star's core will be exhausted, and
conversion to helium will commence in a shell surrounding the core.
Without a nuclear energy source and accreting matter, the core will
slowly contract, increasing the density and temperature of the helium
core. Note that the composition has been radically changed from that
of originally being nearly all hydrogen. The contraction and heating
of the core will cause the outer layers to expand and cool, so that
the star will move to the upper right in the H-R diagram and become
a red giant.
How far a star progresses
past the red giant phase depends upon how much mass it has. Most stars
have enough mass so that the temperature and density of the core increases
until helium begins to fuse in the triple alpha process, called such
because three helium nuclei (alpha particles) come together to form
a carbon nucleus. With a renewed nuclear energy source, the contraction
of the core is reversed for a while and the outer layers are restructured
so that the star moves back toward the left in the H-R diagram to a
horizontal branch above the MS. Eventually the helium in the core is
exhausted, and hydrogen to helium and helium to carbon fusion occurs
in shells concentric with the carbon core. The process of core contraction
is repeated and the star once again swells to a red giant along a path
called the asymptotic giant branch. The most massive stars may pass
through successive steps of fusing helium nuclei with increasingly more
massive nuclei up to iron. Beyond iron, fusion reactions are generally
endothermic (requiring energy) and so cannot be tapped as a fuel source
for stars. Note that these transitions have not actually been observed.
However, they are based on physics principles and will naturally occur.
The lifetimes and rates of change of star changes might be a fruitful
are for creationists to challenge current models.
About 25 years ago it
was realized from the models that the cores of some of the giant stars
should be electron degenerate, which suggested that these stars may
be progenitors of white dwarfs. Degeneracy arises with great density,
when electrons move freely in the stellar core. The atomic nuclei themselves
are tightly spaced in a regular crystalline lattice-like arrangement.
Twenty years earlier it had been demonstrated that electron degeneracy
pressure could account for the very compact structure of white dwarfs
(Chandrasekhar, 1939). It was felt that if the outer layer could somehow
be ejected, then the core left behind would be a white dwarf and we
would have a plausible explanation for their origin. Observations of
red giants reveal that they experience large outward "winds"
that can cause large mass loss over some time. Furthermore the relatively
small surface gravity of red giants would allow for any instabilities
to remove mass at a great rate. The exact mechanism is not quite known,
but most astronomers agree that a red giant can eject a large amount
of mass at some point, leaving behind a white dwarf that is surrounded
by the ejected gas that we would see as a planetary nebula. In the past
25 years a picture has emerged for the most massive stars: their electron
degenerate cores exceed the upper limit (1.4 solar masses) that electron
degeneracy can support, and the resultant collapse to a neutron star
or black hole gives rise to the explosion of a type II supernova.
Observational Predictions of Stellar Evolution
The theory of stellar evolution only briefly summarized above can be
used to make some predictions that can be tested by observations. A
back of envelope calculation can be done to determine how long a star
will remain on the MS is the assumptions above are correct. It is generally
assumed that about 10 percent of a star's mass is in the core and hence
available for nuclear processing. Most of the material (more than 70
percent) will be hydrogen, and we know that 0.007 of the mass will be
converted to energy when hydrogen is fused into helium. Using the mass
and luminosity of the sun yields a MS lifetime of 10 billion years.
More massive stars are higher on the MS, and though they have more mass
available for fuel, their luminosities are are so much greater that
their lifetimes are significantly less. Less massive stars have somewhat
less fuel, but their luminosities are much less, so that their MS lifetimes
are much longer. The upshot is that the most massive stars have MS lifetimes
of only a few hundred thousand years (of course, still much longer than
young-age creationists would allow), while the lowest mass stars have
MS lifetimes approaching 100 billions years.
Suppose that we consider a group of stars that form from a cloud of
gas at about the same time, but having different masses. Because the
cloud we be expected to be thoroughly mixed, the stars would have about
the same composition. Exactly where the MS for such a group of stars
occurs on the H-R diagram depends on the composition: a high metal abundance
(elements heavier than helium in astronomy parlance) shifts it slightly
to the upper right, while a low metal abundance shifts it slightly to
the lower left. For a given composition, the locus of all masses where
stars first appear on the MS is called the zero age main sequence (ZAMS).
As stars "age" on the MS they will hook slightly upward to
the right from the ZAMS, as shown in Figure 2. This is caused by the
change in composition from the thermonuclear reactions occurring in
the core. Upon the exhaustion of hydrogen in the core the star will
move into the red giant region. Because of the greatly increased energy
requirements of a giant star and the diminishing efficiency of post
MS nuclear reactions, the lifetime of the giant phase is only a small
fraction of the MS lifetime.
Figure 2: Theoretical path of stars from the Zero Age Main Sequence
through the Main Sequence band.
Notice that because the upper MS stars have the shortest lifetimes,
they will be the first to turn off from the main sequence. The point
at which this occurs is called the turn-off point (TOP), and it will
be located progressively lower levels on the HR diagram with the passage
of time. Therefore if a collection of stars having the same age and
composition is compared to a different collection of stars having another
common age and composition, the theoretical predictions can be compared
to reality. If one is convinced of the basic correctness of the theory,
then this method can be used to probe the ages of star clusters.
Where do we find stars of the same age and composition? An evolutionary
assumption concludes that the stars in a star cluster should form from
a single cloud so that the members represent such a homogeneous group.
Different clusters should have different ages and though they technically
have different compositions, even large differences in composition do
not seriously affect the overall appearance of an H-R diagram. Generally,
the observations of stars in a cluster consist of colors and magnitudes,
which must be converted to luminosities and temperatures. This process
involves steps such as estimating the distance to the cluster, correcting
for interstellar absorption and reddening from dust, and considering
the effects of stellar atmospheres. This process has been done for a
number of clusters, and the agreement of the theory is quite impressive,
though one wonders how much the theory has been guided along by knowledge
of the data to be fitted. Figure 3 shows schematic diagrams for a "very
young" (100 million years) open cluster and a "very old"
(15 billion years) globular cluster. Note that the globular cluster
has well defined red giant and horizontal branches. This is because
globular clusters contain larger numbers of stars than open clusters
and have TOP's in a region that shows these features well.
Figure 3. Schematic Hertzsprung-Russell diagram for a very "young"
open cluster (A) and an "old" globular cluster (B). On each
graph the zero age main sequence is indicated by a solid line and the
turn off point by the number 1. On the globular cluster plot the Red
Giant branch is indicated by the number 2 an the horizontal branch by
the number 3. Note that the turn off point for the open cluster is much
higher, and that its stars at the lower end lie above the main sequence.
Globular clusters are generally believed to be of about the same age
(15 billion years) while open clusters are believed to have a much wider
dispersion of ages up to 6 or 7 billion years. Additional arguments
for the relative ages of the two types of clusters stem from kinematic
and abundance studies, which spring form a general scenario of galaxy
formation and history, which is an evolutionary model in itself. Stars
that are now old formed early when gas was not confined to the galactic
plane, while younger stars formed after gas collapsed to form the disk.
Thus open clusters are found in the galactic disk and globulars are
formed in the halo of the galaxy. Because of the chemical enrichment
of the interstellar medium that occurs when stars expel processed material
when ejecting a planetary nebula or during the eruption of a supernova,
stars that formed early would be expected to have low metal abundance,
while later stars should have higher metal abundances. Such a trend
between globular and open clusters is observed.
Evidence that the formation of planetary nebulae and the evolution
of white dwarfs are related is usually given in the correlation of the
estimated ages of those two types of objects. The structure of white
dwarfs show that they cool over time, rapidly at first and more slowly
as time progresses, and the rate is very similar for all white dwarfs
so that the temperature roughly reveals the age. Spectroscopic measurements
of a planetary nebula reveal how rapidly the gases in the nebula are
expanding. If the size of the nebula can be measured, then the expansion
can be extrapolated into the past to roughly reveal the age. These two
ages have a very good correlation, that is, the younger planetary nebula
are associated with the younger white dwarfs, and the older planetary
nebula are associated with somewhat older white dwarfs. All planetary
nebula seem to have a white dwarf at the center, but not all white dwarfs
are surrounded by a planetary nebula. How can this be? Planetary nebula
exist for only a brief time before the gases of which they consist are
dispersed into the interstellar medium. The oldest planetary nebula
have estimated ages of only a few tens of thousands of years. On the
other hand, white dwarfs last a very long time, virtually forever. So
the white dwarfs that do not have an associated planetary nebula are
simply old enough to have lost the nebula.
A similar relationship holds for neutron stars and supernova remnants.
As with planetary nebula, the expansion velocity and observed size of
the remnant can be used to estimate the time since the explosion. For
example, the Crab Nebula has an age of about 950 years, and it has the
same position of a supernova observed in the year 1054. The explosion
of a supernova is believed to usually result in a neutron star, which
is normally too small to be ordinarily observed. However, neutron stars
generally emit beams of light from their magnetic poles, and the very
rapid rotation of the tiny star causes a sort of search light of radiation
to sweep out a conical shape. It we happen to lie near that cone, we
can observe the periodic flashes of light, and the star is called a
pulsar. Even with the seeming improbability of being situated near the
cone, more than 300 pulsars are known, with periods on the order of
a millisecond to a few seconds. The rotational kinetic energy is the
source of energy for the beam, and so the period must increase with
time. The rate of change in the period normalized to the period, ,
is directly related to the pulsar's age. Where a pulsar can be identified
in a supernova remnant, the ages of the remnant and the pulsar are well
Very Brief discussions of stellar structure and evolution have been
presented. Though it would seem that creationists would not have much
with which to quarrel in the former, most would largely dismiss the
latter. However, the two are intimately related, and one cannot be rejected
without seriously calling into question the other. We are appealing
to readers to give much attention to the study of stellar evolution,
and we hope that much lively discussion follows.
A number of issues must be addressed. One of the most important is
the question of how much change in stars should a creationist be willing
to grant. Absolutely no change does not seem to agree with obvious observations.
On the other hand, acceptance of most of the stellar evolution with
its required vast ages is not acceptable. A related issue involves the
time scales and rates of change which results from stellar models. These
figures are certainly open to question, but the task is not simple.
If competing creation models are to be presented, then sophisticated,
original computer work lies ahead. Finally, the question arises upon
what physical grounds will rejection of any part of stellar structure
or evolution be done. This part requires that an alternative be offered
for each part, the sum of which should provide some guidance toward
a creationist astronomy.
This is indeed a daunting task, one not to be taken lightly. We plan
to publish further papers, but we certainly welcome correspondence from
interested parties in the meantime.
Chandrasekhar, S. 1938. An introduction to the study of stellar structure.
Dover. New York.
Clayton, Donald D. 1968. Principles of stellar evolution and nucleosynthesis.
McGraw-Hill. New York.
Cox, John P. and R. Thomas Giuli. 1968. Principles of stellar structure.
Gordon and Breach. New York.
DeYoung, Don B. and David E. Rush. 1989. Is the Sun an age indicator?
_____ and John C. Whitcomb. 1981. The origin of the Universe. CRSQ 18:83-90.
Gish, Duane T. 1975. A decade of creationist research. CRSQ 12:34-46.
_____ . 1989a. More creationist research (14 years) - part I: geological
research. CRSQ 25:161-170.
_____ . 1989b. More creationist research (14 years) - part II: biological
research. CRSQ 26:5-12.
Humphreys, D. Russell. 1990. Good news from Neptune: The Voyager 2 magnetic
measurements. CRSQ 26:15-17.
Iben, Icko. 1991. Single and binary star evolution. Astrophysical
Journal (Supplement). 76:55-114.
Kippenhahn, R. and A. Wiegert. 1990. Stellar structure and evolution.
Springer-Verlag. New York.
Mulfinger, George, Jr. 1970. Critique of stellar evolution. CRSQ 7:7-24.
_____ , Editor. 1983. Design and origins in astronomy. Creation Research
Society Books. Kansas City, MO.
_____ . 1973. Review of creationist astronomy. CRSQ 10:170-175.
Novotny, Eva. 1973. Introduction to stellar atmospheres and interiors.
Oxford University Press. New York.
Schwarzschild, Martin. 1958. Structure and evolution of the stars. Dover.
Wilt, Paul. 1983. Nucleosynthesis in Mulfinger, George, Jr. (editor).
Design and origins in astronomy. CRS Books. Kansas City, MO.