The laws of physics do not necessarily break down at the event horizon of a black hole. Rather, think of it like this.
Before Newton, people got along just fine and thought they knew a lot about how the universe worked. When Newton (and Leibniz) developed calculus, we finally had a system of mathematics rich enough to describe the kinematics and dynamics of everyday objects in motion as well as heavenly bodies. It's not that the universe didn't obey the laws of physics before then, it's just that we had no way of adequately describing them. Once we had calculus, it became possible to make accurate predictions regarding these phenomena. It was also possible to begin unifying what had previously been independent areas of research (e.g. ballistics and astronomy).
Likewise, the universe has always behaved in strange ways (relative to our common sense) at very small length scales, very large mass/energy densities, and velocities approaching the speed of light. However, before we had quantum mechanics and the special and general theories of relativity, we just didn't have any way of properly thinking about this weird behavior, nor any way to properly observe it. We arrived at these advanced theories, not because they naturally flowed from Newton's laws of motion or Maxwell's laws of electromagnetism, but because physicists began noticing that their experiments were not returning the results they expected. When the theory makes predictions that are not supported by observed facts, then you have exceeded the domain of validity of the theoretical model. To proceed any further, someone has to develop a more sophisticated model which not only accounts for the new observational facts, but also simplifies to the previous model when considering phenomena for which the previous model was making accurate predictions. Quantum mechanics and the theories of relativity do this with respect to Newton's and Maxwell's laws.
Now, with respect to black holes, there is an analogous situation. Here we do not have sufficient experimental data to develop a new model, however, the predictions made by existing models are not completely satisfactory either. In particular, when quantum mechanics and relativity are applied to the problem of black holes, it is difficult to arrive at a clear picture of what to expect. The theories provide somewhat incompatible predictions. Ordinarily, this disagreement would be settled in the laboratory. May the theory which makes the most correct predictions win. Unfortunately, there are not that many conveniently located black holes for us to study, nor can we easily recreate the physical conditions under which these phenomena would be observable. That may be about to change, however, once the Large Hadron Collider becomes operational. It is said that the LHC might be capable of producing sufficient mass/energy density that a micro black hole could possibly form, and then instantly evaporate. If such an event does occur, it would be a tremendous opportunity for physicists to begin answering some of these long held questions regarding the uneasy interface between quantum mechanics and relativity.
Long story short: Black holes are currently beyond the well established domains of validity of our current theoretical models. This situation will most likely only be remedied by one of two possibilities. One possibility is that we begin getting some hard data on black holes from the LHC or maybe from more advanced telescopes. The other possibility is that somebody eventually comes up with an elegant mathematical framework which describes not only all of our current laws of physics, but also is able to generate unambiguous predictions in the regime where quantum mechanics and relativity meet. Of course, such a theory must also be able to stand up to independent experimental verification. Otherwise it's just a model and may or may not accurately reflect reality (e.g. string theory).