Some authors have defended the claim that one needs to be able to define ‘physical coordinate systems’ and ‘observables’ in order to make sense of general relativity. Moreover, in Rovelli (Physical Review D, 65(4), 044017 2002), Rovelli proposes a way of implementing these ideas by making use of a system of satellites that allows defining a set of ‘physical coordinates’, the GPS coordinates. In this article I oppose these views in four ways. First, I defend an alternative way of understanding general relativity which implies that we have a perfectly fine interpretation of the models of the theory even in the absence of ‘physical coordinate systems’. Second, I analyze and challenge the motivations behind the ‘observable’ view. Third, I analyze Rovelli’s proposal and I conclude that it does not allow extracting any physical information from our models that wasn’t available before. Fourth, I draw an analogy between general relativistic spacetimes and Newtonian spacetimes, which allows me to argue that as ‘physical observables’ are not needed in Newtonian spacetime, then neither are they in general relativity. In this sense, I conclude that the ‘observable’ view of general relativity is unmotivated.

一些作者为这样的说法进行了辩护，即需要能够定义“物理坐标系”和“可观察对象”才能理解广义相对论。此外，在 Rovelli （Physical Review D，65（4）， 044017 2002） 中，Rovelli 提出了一种利用卫星系统来实现这些想法的方法，该系统允许定义一组“物理坐标”，即 GPS 坐标。在本文中，我从四个方面反对这些观点。首先，我捍卫一种理解广义相对论的另一种方式，这意味着即使没有“物理坐标系”，我们对理论模型也有完美的解释。其次，我分析并质疑“可观察”视图背后的动机。第三，我分析了 Rovelli 的提议，并得出结论，它不允许从我们的模型中提取任何以前不可用的物理信息。第四，我在广义相对论时空和牛顿时空之间做了一个类比，这使我能够论证，既然牛顿时空不需要“物理可观察物”，那么它们在广义相对论中也不需要。从这个意义上说，我得出结论，广义相对论的“可观察”观点是没有动机的。