I don't see what the issue is.
I think it's fairly certain that the Barnett who appears as 'married' to Louisa for 23 years in the 1911 census is 'our' Barnett. 23 years puts the start of their relationship in 1888, if we accept that as a true reckoning.
Do their individual claims to be married mean that they were actually officially married? Of course not. Many people who were not officially married stated that they were in the census return. It's a matter of respectability as much as anything. What it doesn't do is confirm that their was ever an official marriage - so we may be wasting our time in looking for one.
I don't know where the identification with Louisa Rowe came from in the first place - I note that a couple of family trees on Ancestry cite her as the wife of 'our' Barnett; however, I don't recall ever seeing any supporting documentation, and it may well amount to a guess based on the only recorded marriage a Joseph Barnett to a Louisa at roughly the right time being that of Joseph Barnett to Louisa Rowe.
When Chris said that the details on the marriage certificate didn't match; he'd have compared the known details regarding Barnett's life that would've been recorded on the marriage certificate; like age, occupation and father's name and occupation. If these details don't match, it isn't the right man.
As mentioned earlier, there is another Joseph an Louisa Barnett who I think I concluded a few years ago were the people married in 1887. As I can no longer find my notes from the time, I can't tell you exactly how I came to that conclusion, so I'll either have to try to retrace my steps or order the marriage certificate myself. The latter might prove easier.

I hope that helps to clarify things.
Leave a comment: