Draft version June 28, 2012 A Preprint typeset using LTEX style emulateapj v. 5/2/11 MODERN COSMOLOGY: ASSUMPTIONS AND LIMITS Jai-chan Hwang1,2 1Department of Astronomy and Atmospheric Sciences, Kyungpook National University, Daegu 702-701, Republic of Korea 2Korea Institute for Advanced Study, Seoul 130-722, Republic of Korea Draft version June 28, 2012 ABSTRACT Physical cosmology tries to understand the Universe at large with its origin and evolution. Ob- servational and experimental situations in cosmology do not allow us to proceed purely based on the empirical means. We examine in which sense our cosmological assumptions in fact have shaped our current cosmological worldview with consequent inevitable limits. Cosmology, as other branches of science and knowledge, is a construct of human imagination reflecting the popular belief system of the era. The question at issue deserves further philosophic discussions. In Whitehead’s words, “philosophy, in one of its functions, is the critic of cosmologies.” (Whitehead 1925) 1. SCIENCE AND COSMOLOGY by the external world.” Alexander Calder has mentioned that “the universe Although science is an effort to approximate phenom- is real but you can’t see it, you have to imagine it.” ena through models and to test those models, it is im- As an artist Calder’s universe may mean everything in portant to be aware of the difference between model and the world. Modern physical cosmology aims at under- reality. As remarked by Alfred North Whitehead, “the standing quantitatively the structure, origin and evolu- aim of science is to seek the simplest explanations of com- tion (sometimes including future) of the Universe based plex facts. We are apt to fall into the error of thinking on scientific methods (i.e., observation and experiment). that the facts are simple because simplicity is the goal of our quest.” (Whitehead 1920) To scientists, the Universe is a physical object1 in the large-scale. However, regions in the Universe accessible 2. by observation and experiment are limited. The forbid- COSMOLOGICAL PRINCIPLE den regions include outside our current cosmic horizon, In cosmology, in order to infer regions unknowable even time already passed in our past light-cone, and the far in principle, an assumption is necessary, which might future which has yet to come. Some of these regions are reflect our anticipation. An assumption that distribu- not just practically difficult to access, but they belong tion of matter in the large-scale is spatially homogeneous to the absolute limit of scientific knowledge which is in and isotropic is termed the cosmological principle, often principle inaccessible by observation or experiment. adopted in modern physical cosmology. Its origin can Of course, science is not simply based on observation be traced to Einstein’s paper in 1917, in which he has and experiment so innocently. If science were naively applied his newly introduced gravity theory to cosmol- based on observation and experiment, the science as we ogy in order to reconcile his theory with Mach’s principle know now may not be possible. On the contrary, science (Torretti 2000). Although even the presence of external in reality is more related with the art of ignoring and se- galaxies was not known at that time, Einstein assumed lecting observations, and manipulating experiments, in homogeneity and isotropy of space, and thus of the mat- accordance with a preconceived theory. Detailed obser- ter distribution, merely for the sake of mathematical sim- vation is often a hindrance to scientific reasoning. Ignore plicity. Notice that this is an assumption not based on apparent phenomena and grasp the essence. Thus, in what was observed. Perhaps Einstein did not expect that science theory often comes before observation. The trick this simple working assumption would have become a ba- is to treat the subject as an isolated, simplified, idealized sic principle in cosmology in future development. and abstract (preferably mathematized) model, and to In fact, the basic tenet of physics is that the laws of arXiv:1206.6297v1 [physics.hist-ph] 25 Jun 2012 test and materialize it by fitting data to a model using physics we know are valid always and everywhere. Thus, the method of analysis and statistical techniques. In this advocating the universality of the laws of physics reminds way, the individuality is lost. Modern cosmology shows us sort of the cosmological principle; this belief in univer- such a state of affairs well. After all, scientific cosmology sality is in general not testable. It is amusing to notice is nothing other than ‘the Universe imagined based on that assuming the general validity of the physical laws the “scientific” method.’ In fact, the situation of cosmol- in space and time is more similar to the perfect cosmo- ogy is not alone in science. In Albert Einstein’s words, logical principle, which adds the time invariance of the “physical concepts are free creations of the human mind, physical state of the Universe in addition to the cosmo- and are not, however it may seem, uniquely determined logical principle. However, the perfect cosmological prin- ciple implemented in the steady-state theory is no longer popular in modern cosmology. 1 Professor Roberto Torretti objects calling the Universe as a “physical object”. According to him, “the epistemic counterpart Although in history the cosmological principle was ini- to artist Calder’s dictum would be something like: ‘The universe tially postulated theoretically without any reference to is real, but you can’t grasp it as an object, you have to think of it observations, one might wonder whether the assumption as a Kantian Idea’.” (Torretti 2011) on matter distribution can be tested through the ob- 2 servations. However, there are difficulties in practice, any potential dependence of the rate on the angular di- and often in principle. There can be some evidence of rection of the object in the sky. This is surely due to isotropy around us, but the test of homogeneity becomes the influence of the cosmological principle. Thus, even difficult as the scale increases. In addition, if we con- the observation is performed under direct influence of the sider the finite speed of light, even in a perfect observa- theory. In this case, the theory is nothing more than our tion, we cannot prove the homogeneity of space. Only assumed cosmological principle. through models we can agree on its plausibility. The ob- Motions of galaxies in clusters and the rotation speed served two-dimensional projected isotropy of the cosmic of disks in spiral galaxies are known to be too fast to microwave background radiation does not necessarily im- be bounded by luminous matter. Without substantial ply that the three-dimensional matter distribution is also amount of non-luminous matter present, galaxies and isotropic. Furthermore, examination beyond the horizon clusters are unstable and could be transient phenom- (light propagation distance during the age of the Uni- ena. Such non-luminous matter, only known through verse) is in principle impossible, and the cosmological gravity, is termed as ‘dark matter’. Such an interpreta- principle in those regions remains as an untestable as- tion, however, is based on the assumption that Newton’s sumption. As emphasized by George F. R. Ellis, “the (Einstein’s as well) gravity is valid in galactic and cluster problem [is that] there is only one universe to be ob- scales. This reflects our belief in the universality of phys- served, and we effectively can only observe it from one ical laws, and particularly our faith in Newton’s theory, space-time point. Given this situation, we are unable to which lead us to such a conclusion. But Newton’s the- obtain a model of the Universe without some specifically ory has never been tested on those scales; nor Einstein’s cosmological assumptions which are completely unverifi- gravity has yet been tested in cosmology. Therefore, dark able.” (Ellis 1975) In modern cosmology, the assumption matter is also a case where our belief system has affected refers to the cosmological principle. the interpretation of the observed results. Theoretically, without the cosmological principle, Einstein’s gravity is widely accepted as the gravity to physical cosmology becomes mathematically too compli- handle astronomical phenomena. The theory holds a re- cated to handle, and this practical difficulty might have markable track record in the solar-system test based on an important role in accepting this simple assumption. vacuum Schwarzschild solution and the parameterized Martin Rees has mentioned that “principles in cosmol- post-Newtonian approximation where the gravitational ogy have often connoted assumptions unsupported by fields are supposed to be weak. Although it is true that evidence, but without which the subject can make no Einstein’s theory has not failed in any experimental test progress.” based on modern scientific and technological develop- The cosmological principle still in large measure is a ment up till today, it is also true that there has been philosophical assumption, and is not based on obser- no experimental test of the theory in the strong gravita- vations or experiments as is often emphasized by the tional field and in large scale even including the galac- scientific method. Making assumptions is not a prob- tic scale. Cosmological application of Einstein’s theory lem. It is fine, as long as we are aware of this nature, requires 1015 factor (horizon scale divided by an astro- and try to examine the case in regions where testing is nomical unit) extrapolation compared with the experi- possible. From the observational side, efforts on tests mentally tested scale, which is surely a staggering ex- of whether the Universe is homogeneous (homogeneity trapolation. Einstein’s gravity is generally accepted in measure) and isotropic (isotropy measure), and whether cosmology mainly based on its successes in other astro- deviations from the homogeneity-isotropy are acceptable nomical and Earth bound tests and the theory’s own (linearity measure), should be continued.