I think we might give you an infinite variety of answers, because each of us had a difference experience, different backgrounds and stories to tell. IMHO, the question itself has a serious problem: why choosing? Do you choose between a hammer and a nail to put a frame on the wall? No, because both are tools, as Python and R. Obviously, you would start learning only one of those but in the end, I would suggest learning both of them because, at least today, they are complementary.
In my experience (scientific background, working with energy and climate data), Python is very good in handling large gridded datasets thanks to
xarray and in performing fast computation, on the other hand I use R when I need the "last mile", i.e. when I have to process the data to visualize it (
ggplot2 to day is the top library for data exploration and visualization) and to present it (the
shiny library is very very easy to learn and effective). But while R syntax can be ugly and not consistent, Python is very elegant and definitely a modern programming language, the best solution when you need to write a lot of code with a OOP paradigm.
So the question would be: which one I should learn first? It depends: why are you learning it? Do you have already a project? If you still don't have a clear idea, I would suggest attending the first R and Python courses in Datacamp and then choosing where to start.