Interesting point. I suppose a "law" is usually an empirically verified pattern or relationship in nature, often quantitative and frequently given the name of the person who enunciated it. So it can be seen either as a summary of observations (which are a matter or fact) or as the first step towards building a theory out of a set of observations. But even "laws" can be shown to be false or only approximations. After all Newton's 2nd law does not survive contact with relativity - or at least has to be modified (re-expressed in terms of relativistic momentum) in order to do so. And Newton's law of gravitation is shown to be inaccurate, and is completely replaced, in GR.