Dr. Matravers et al., COMPLEMENTARY APPROACHES TO COSMOLOGY - RELATING THEORY AND OBSERVATIONS, Quarterly Journal of the Royal Astronomical Society, 36(1), 1995, pp. 29-45
The usual way of proceeding to determine the structure and evolution o
f our observable Universe is to assume that on the largest scales it i
s Friedman-Lemaitre-Robertson-Walker (FLRW), i.e. isotropic and spatia
lly homogeneous, and then to determine the parameters specifying the F
LRW model from observations. However, in the light of observational ev
idence of deviation from a FLRW geometry on almost all scales, cosmolo
gy must gradually clarify, and if necessary modify, the geometric assu
mptions upon which this procedure rests, along with other presuppositi
ons, such as the fluid-character of the mass-energy content. The basic
requirement is to balance the two major approaches to cosmology: the
FLRW model-building approach, starting from broad explanatory premises
, and a descriptive approach, starting on a strictly observational bas
is. Each gives us different insights and information about the nature
of the Universe. Lacking a model that integrates both satisfactorily,
we need to develop each approach as far as it will go, repeatedly comp
aring the two and seeing what each can teach the other. As a particula
r example, inter alia observational cosmology eventually needs to spec
ify the length-scale at which the Universe becomes FLRW or 'almost FLR
W'. This can only be done by pursuing both approaches. Here we review
briefly the successes of the standard approach and also its limitation
s, and indicate how we should go beyond it to develop a fully satisfac
tory understanding. We describe briefly one particular integrated appr
oach as an example of what can be done, assess its feasibility, and re
late it as an essential complement to the standard FLRW programme.