I found it tiring that she consistently dismisses the ancient sources as wrong without giving an reason why. Academics have been saying "well this <commonly believed thing from antiquity> can't possibly be true", and it not infrequently turns out to be at least not wrong. Sure, sources in antiquity are unreliable, but they are at least two millennia closer to the actual events than we are.
The book also seemed like a survey that had chunks missing out of the map. The parts that were there were well done (excepting not giving any good reason for distrusting the ancient sources) and I definitely learned things. But I don't feel like I got a good feeling of the course of Roman history, more like little vignettes along the way.
I can't say I dislike the book, but I think it could be better. (Although, I'm a complete novice in Roman history, so that might influence my opinion. And maybe the book wasn't written for me, too.)
I read it a few years ago after listening to the Duncan podcast and reading a bunch of archeology papers. I'm definitely no classicist, and don't read Latin but have a passing interest in the history. I think some of here dismissals are well founded and other experts agree. I think the weakness of the book is that it leans heavily on pronouncement of unknowability. But I really like that she treats the reader like a smart person and presents conflicting and complicated information as is, and lets you mull it over.
It isn't a perfect book, but it's very readable, informative, and seems to be well regarded. I was on a Rome kick at the time and didn't find anything that seemed to be completely out of left field.
Perhaps it's a reaction to the normal bias? I'm constantly surprised by how people will take extremely partial accounts from somebody on one side of an issue and no real expressed or implicit commitment to accuracy at face value.
[0]: e.g. Thucydides' History of the Pelopenesian war might be accurate, but honestly, why would it be? It's even worse when you have stuff like Caesar's writings.
