I'm not so sure UX can be dismissed that easily. IA deals with content hierarchies rather than things like "navigation efficiency" details or site registration journeys, or touch interaction behaviours and optimisation.
There's cross-over, but one "Designer" can't do it all, even if they have a good grasp of everything.
When the focus is on UX, decisions might be made about how people sign in or out. The UX expert might determine that a 2-step log out process is one step too many, but the research they gather might suggest it's nevertheless still acceptable enough to get away with. The designer makes whatever-step process look and function the best it can, without needing to scrutinise the research and testing around 2-step sign out processes.
Another example - consider the user experience of dismissing the banner that appears on some websites that have an app in iOS Safari. The little 'X' close link is tiny, way smaller than any touch button or link should be. Bad UX? Sure is. But the UX person must have data suggesting that having a stupidly small close button for that banner is worth it for the primary visual incentive to act on the promotional prompt without annoying too many people (annoys the hell out of me, but I'm not everyone).
There's cross-over, but one "Designer" can't do it all, even if they have a good grasp of everything.
When the focus is on UX, decisions might be made about how people sign in or out. The UX expert might determine that a 2-step log out process is one step too many, but the research they gather might suggest it's nevertheless still acceptable enough to get away with. The designer makes whatever-step process look and function the best it can, without needing to scrutinise the research and testing around 2-step sign out processes.
Another example - consider the user experience of dismissing the banner that appears on some websites that have an app in iOS Safari. The little 'X' close link is tiny, way smaller than any touch button or link should be. Bad UX? Sure is. But the UX person must have data suggesting that having a stupidly small close button for that banner is worth it for the primary visual incentive to act on the promotional prompt without annoying too many people (annoys the hell out of me, but I'm not everyone).