If the manufacturer does not provide the sequence (and most do not because that information is typically proprietary) they may be willing to provide an approximate region to help you narrow it down -- just ask. Then you will still have to perform epitope mapping experiments to narrow it down to the resolution you require. I've not performed these experiments myself but I believe they are non-trivial.

However, that's assuming they used a conjugated peptide sequence for antibody development -- this is almost always true for an antibody that targets a PTM, but in other cases they may have used a recombinant protein (fragment or even full-length). If they did NOT use a specific peptide sequence to raise the antibody then it's very likely that even the manufacturer does not know where the binding epitope is, and epitope mapping experiments become even more complicated. 

You should ask yourself how important it is to know the epitope, and what granularity you need (i.e. do you need the precise epitope or is an approximation enough?). You may be able to narrow it down by combining it with other antibodies to your target that DO have known binding epitopes.

Thank you, William!

As William correctly mentioned, most of the antibody manufacturers or suppliers either do not know the bound epitope or do not want to disclose it. I think that this is an irrational behavior since they damage their own customers. In the linked article, you can read more about these issues. I also agree that a real epitope scan is a big effort and is often not necessary for a practical application.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4803150/

If you really want to reproduce exactly the same epitope (which is not always necessary) you could select phage-displayed peptides on the immobilized antibody. Peptides thus selected from a random library can contain motifs recognized by the antibody.  This is quite easy, but these motifs could resemble or not the original epitope. There are examples of  perfect mimicry and also failures. You can try, testing the antibodies raised against the selected petides (or peptide motifs) with the original antigen. If they react, you are lucky and the specificity was copied.