Do Black Holes Exist?

Proof that black holes exist will likely require confirmation of the existence of event horizons. The common assumption that the mere existence of large compact masses proves the case for black holes is an unwarranted extrapolation of General Relativity into a strong-field regime where it has not been adequately tested. Neither the large compact masses of galactic nuclei nor the massive compact objects of stellar mass in the x-ray binaries prove the existence of black holes. In contrast to the case for galactic nuclei, we have the necessary tools for obtaining either proof or disproof of event horizons in the x-ray binaries. Observations of kHz QPOs may decide the event horizon issue very quickly. If not, we can still obtain proof by comparing predictions of gravity theories that differ primarily by the presence or absence of an event horizon. Detailed analysis of models of x-ray binaries would then decide the issue. A preliminary analysis given here, based on the Yilmaz modifications of General Relativity, indicates that it may already be decided. The spectral properties of x-ray binaries are consistent with the presence of surfaces rather than event horizons. Prominent features such as hard power-law tails of the x-ray spectrum, spectral state switching, slow QPOs, flickering, ultrasoft peaks, hard lags and strong coherence of hard and soft emissions have natural explanations as surface phenomena. Ultrasoft peaks appear to be the signatures of strongly redshifted boundary layer emissions. Current black hole plus accretion disk models explain few of these features. They are also inconsistent with time resolved spectra