A fundamental question in cosmology is, "How did the universe begin?" The two pivotal ideas of infl ation and cold dark matter (CDM), combined with extensive observational results, including the unpredicted accelerated expansion of the universe, underpin a new standard model of cosmology. Modern cosmology began with the application of general relativity to interpret the relation between galaxy redshifts (how fast they are speeding away) and their distance from us ( 1). The presence of large

more »

... nts of dark matter was deduced early on ( 2) and has become part of the standard model. The discovery of the cosmic microwave background (CMB) ( 3) and its blackbody spectrum ( 4, 5) solidifi ed the big bang theory, whereby the present universe resulted from the expansion and cooling of an initially hot and dense environment (see the fi gure). However, the big bang theory only describes the expansion and cooling but says nothing about the origin of the universe. Within the standard model, the beginning of the universe is effectively "infl ation," the rapid expansion of a tiny region of space to astronomical scales ( 6). Infl ation is a paradigm that encompasses a wide range of spe-cifi c implementations that are at the intersection of quantum and gravity theories, the two great but incompatible theories of 20th century physics. Measurements of infl ation not only will probe the origin of the universe but also may help reveal the basic structure of physics itself. The measured amplitude of CMB anisotropy is consistent with the gravitational growth of structure, but only if CDM dominates over baryonic mass (ordinary matter). The fullsky Wilkinson Microwave Anisotropy Probe (WMAP) data strongly constrain cosmology, especially when combined with other cosmological data: baryon acoustic oscillation ( 7) , Hubble constant ( 8) , and type Ia supernovae ( 9). This data combination is the basis of the standard model ( 10) , a precise quantitative description of a universe dominated by a constant energy density (driving the accelerated expansion), supplemented by cold dark matter and atoms, with density fl uctuations from infl ation. The constant energy density may be "vacuum energy"-strangely, space can be "empty" but still contain energy in the form of a negative pressure that drives the expansion. Type Ia supernovae are the best way to measure accelerated expansion at small redshifts. Current results are limited by systematic effects that need to be addressed for continued progress to be made. Much larger cos-mic volumes at larger redshifts favor methods that make use of an enormous number of galaxies: weak gravitational lensing ( 11) and baryonic acoustic oscillations ( 12) . The inability to predict the level of the vacuum energy predates the detection of cosmic acceleration. With great effort and considerable resources, cosmologists can make only about an order of magnitude improvement in measurements to attempt to falsify the null hypothesis: that vacuum energy is the cause of the cosmic acceleration. Fortunately, many observational approaches also provide critical data for the other key advances as well. There are six generic predictions of the simplest versions of inflation: (i) the temperature of the CMB should be highly uniform across the sky, as verifi ed by the Cosmic Background Explorer COBE; (ii) the universe should have a nearly, but not precisely, fl at Euclidean geometry. The fi rst indication of fl atness ( 13) has progressed to the current measurement that the Universe is fl at to ~1% accuracy; (iii) the temperature fl uctuations should display only a slight deviation from the preinfl ation conjecture of equal power on all angular scales, and just such a small deviation is now observed with nearly 3σ significance; (iv) infl ation should have driven adiabatic fl uctuations to superhorizon scales, and these should have later re-entered the hori-