You don't know what you are talking about either. CDMA »www.cdg.org/resources/cdma_history.asp was first conceptualized in 1988, ratified as a standard in 1993, and the first networks went into service around that same time. EVDO was developed in the late 90s (97-99 timeframe), and is still here 15 years later.
So, really, us Americans build something and "it becomes useless down the track"? What a load. I'd say a 20 year old technology held up pretty good, considering it's probably going to be a 30 year old technology by the time it's retired from voice use. But it's not European, so it must be inferior. (hint: LTE is not a European technology, and the first WCDMA networks went into service in Japan in 2001 and were in use in the United States in 2004, before most of Europe).
CDMA has been up and running since the 90s, and with EVDO it kept pace with 3G GSM technologies. People bemoan the max speed of 3.1Mbit/sec on Rev.A (which is higher with newer Protocol Revisions), but that is per sector, per 1.25MHz EVDO channel. A cell site has 3 sectors with multiple channels running. So yeah, the MAX you'll ever get on your handset is 3.1Mbit/sec (darn) , but it's not like one site is only capable of providing 3.1Mbit/sec to all its users. If you look at the spectral efficiency of 5MHz of EVDO (3 EVDO channels, the minimum channel size is 5MHz for one channel of 3G WCDMA), it's somewhere around 9.3Mbit/sec. Not bad for a technology older than UMTS. If you want to get really technical, Rev. B bumps a 5MHz channel up to around 15Mbit/sec and drops latency to around 40ms.
You make it sound like CDMA was installed in 2005 and died on the vine. Nothing is further from the truth.
If your beef with American-developed technologies is that they keep running fine for so long that greedy operators fail to move forward until the last possible second, then say that. But saying that our technologies don't scale really is untrue.