From mathematical perspective chess is solvable in a sense that it does have a solution and there is a well defined algorithm for finding it. Period, end of story.
There are some arguments that it is not solvable practically as we may never have enough computational power. I do not have a s strong opinion on this one.
It's only because computer engine assessments of positions are in common use that some people have been induced to believe that single, best moves exist in all positions, which is incorrect, although, of course, single, best moves do exist in positions with "forced moves" for which an alternative move would bring about a worse game result. However, humans have input into engine programming and also chess has never and almost certainly will never be "solved".
There is not necessarily a single best move in every positions, but there is always a certain number of best moves in positions that are either a forced win or a forced draw for a player making the move. From game theoretical point of view they are all equally strong, all other moves being bad. In a third type of position, in which our opponent has a forced win, there are simply no good moves (or you might say that the set of best moves is empty).
Chess has not been solved, but from a theoretical (mathematical) point of view it is absolutely solvable. We know for a fact that the solution exists, we just may never have enough computing power to find it.
If you didn't visit Ponz's thread then you mightn't believe how much this was debated there. The thread's quiet now because it couldn't go any further.
The question as to whether chess is solvable was raised. The prevailing attitude was that it definitely isn't solvable. That would be from a brute force search, because there are too many possible variations even for the fastest computers to complete it in thousands of years.
I was the lone voice there arguing that chess is potentially solvable from a mathematical perspective. I tried to explain how I thought that new algorithms might be developed that would analyse certain aspects of chess in order to express the unbalancing of a position and achievement of a winning position resulting from that by a mathematical expression. I'm not a mathematician but my son is a professional mathematician and it was watching his progress through his mathematical education that made me understand that maths is still evolving and the methodology is still strengthening. New processes are being invented/discovered. So I believe it's possible.