In this manuscript we initiate a systematic examination of the physical basis for the time concept in cosmology. We discuss and defend the idea that the physical basis of the time concept is necessarily related to physical processes which could conceivably take place among the material constituents available in the universe. As a consequence we motivate the idea that one cannot, in a well-defined manner, speak about time ‘before’ such physical processes were possible, and in particular, the idea that one (...) cannot speak about a time scale ‘before’ scale-setting physical processes were possible. It is common practice to link the concept of cosmic time with a space-time metric set up to describe the universe at large scales, and then define a cosmic time t as what is measured by a comoving standard clock. We want to examine, however, the physical basis for setting up a comoving reference frame and, in particular, what could be meant by a standard clock. For this purpose we introduce the concept of a `core' of a clock and we ask if such a core can - in principle - be found in the available physics contemplated in the various `stages' of the early universe. We find that a first problem arises above the quark-gluon phase transition where there might be no bound systems left, and the concept of a physical length scale to a certain extent disappears. A more serious problem appears above the electroweak phase transition believed to occur at $\sim 10^{-11}$ seconds. At this point the property of mass disappears and it becomes difficult to identify a physical basis for concepts like length scale, energy scale and temperature - which are all intimately linked to the concept of time in modern cosmology. This situation suggests that the concept of a time scale in `very early' universe cosmology lacks a physical basis or, at least, that the time scale will have to be based on speculative new physics. (shrink)