Linear Regression - About R-squared and slope

My question is that why low or high R-squared has no bearing on whether the estimated slope has a causal interpretation. My explanation is because R-square is the estimator of how well x explains the variation of y, therefore R-square has no inference about the causal relationship.
I'm not sure if this explanation is correct and sufficient enough. If possible, could you help elaborate more on this? And an example, may be :D Thanks.