The main part of this section is based on B. J. T. Jones' review J and P. V. B. Lilje's Cand. Real. thesis L.
Cosmogony, the science of the origin of structures in the Universe, is by far no novel science. It has troubled scientists, philosophers, and theologians for at least two and a half thousand years. There might not be much for us to learn from the most ancient works, but some of the more general ideas have validity today. Take the Epicureans for example; they meant that the order we see around us today was once in early history complete chaos. This idea can very well be compared to the turbulence theory, which will be presented below. Aristotle, on the contrary, believed that order could not be created from disorder. Newton supported this idea by arguing against the vortex cosmogony of Descartes, on the ground that it would fail to reproduce the cosmic order displayed by Kepler's laws. These ideas are similar to the gravitational instability theory, which states that entropy increases with time.
Even though the origin of structures was considered an important question, very little
progress was made until the beginning of this century. The first photographs of
galaxies lead to a number of papers on the subject, e.g. Jeans-02, who stated
that a gas cloud exceeding a certain size (the Jeans length) would collapse. He later
brought this idea into cosmology Jeans-28. The big breakthrough was, naturally,
not made until the establishment of extra-galactic distance scales (about 1924) and
Hubble's discovery of the general expansion of the Universe in 1929. The gravitational
instability theory was starting to form. One milestone came with Lema^itre, who
reconciled the general expansion and Einstein's theory of gravity, and argued that the
gravitational instability in a uniform expansion is the origin of galaxies. In the
Lema^itre model Lem, the Universe expands from a dense state, the
``Big Bang'' (the expression was later coined by Fred Hoyle as a derogatory remark), and
passes through a quasistaic or coasting phase where the expansion rate and acceleration
are small because the cosmological constant almost satisfy the conditions of the static
Einstein solution. Another important paper was by Ga-Te, who applied Jeans'
instability to the expanding Universe. Yet another milestone came with Lif. He
showed that in an Einstein-de Sitter model, linear density perturbations grow as
, which was considered to be far too slow. Bon arrived at the
same result, but then by Newtonian gravitational theory. He showed that with the
approximations Lifshitz used ( 
, 
and 
, where
``PH'' denotes the particle horizon), one could just as well use Newtonian theory
instead of general relativity. The results of Lifshitz and Bonner were considered a
serious blow against the gravitational instability theory.
An alternative to the gravitational instability picture, the turbulence theory, saw the
light of day in 1951. W pictured the galaxies as remnants of some primordial
turbulence, a view that may not seem that controversial when considering the shape of
the spiral galaxies. This theory would soon become more attractive than the
gravitational instability theory, mainly due to the problem of the latter of explaining
the fast growth of the irregularities. Even though the turbulence picture now had an
upper hand, it had quite a few problems of its own and was barely considered in the
latter half of the 1960s, until Ozernoi and his co-workers picked it up, then within the
framework of the ``hot Big Bang'' cosmology. This did little to help the theory though;
it predicted density fluctuations that were far to large, and it implied divergent
fluctuations in the spatial geometry for 
. Since the early seventies,
the turbulence theory has rarely been mentioned.
One now had a situation where no theory could explain the formation of galaxies. What saved the situation was the discovery of the 3 K cosmic background radiation field in 1965 by Penzias and Wilson, and the general acceptance of the Big Bang theory. There was no longer any reason why the original perturbations could not have grown into the structures we see today. At the present, there is a general acceptance for the gravitational instability theory.