You can put limits on how much they could have changed, though. And people do this, e.g., [0], which found the proton-to-electron mass ratio has not varied by more than a relative factor of 4e-7, over the past 7.5 billion years.

[0] http://arxiv.org/abs/1412.7757