The iPhone came out before Android, was more expensive than a Blackberry at first, but started taking share from RIM. It also caused the Android project at Google to reset (their initial phone was similar to a Blackberry before the iPhone launch).
RIM’s complacency and flat-footed response to the iPhone has been very well covered, and can be Googled. By the time they were trying to partner with Google, it was already too late for them. Android is not what killed the Blackberry.
Except this assumption that there can be only a few winners taking all and the rest end up "killed" is exactly what effective anti-trust would prevent. Imagine the possibility where Blackberry kept making devices that fit their niche of hardware keyboards, running Android, and selling them to their established customers.
The elephant in the room is the heavy natural monopoly on software compatibility/interoperability, and the above thought experiment only makes sense because Android is a purportedly open platform.
Effective tech anti-trust has to work around that and not merely police behavior aimed at restricting competition. Modern tech anti-trust should be focused around mandating open access to proprietary systems (for starters, anything a web user can do, a program running on the user's behalf should be able to do), and privacy legislation to make it so that customers have to be earned rather than everyone being treated as a data subject by default.
Not reset, reprioritize. Fred Vogelstein only talked to one source who was on the Android team for his book, and that source may have had their own agenda.
Other sources, such as Chet Haas's book, make it clear that what became the G1 was already on the roadmap. It was just prioritized.
RIM’s complacency and flat-footed response to the iPhone has been very well covered, and can be Googled. By the time they were trying to partner with Google, it was already too late for them. Android is not what killed the Blackberry.