After you get into it, I've found that syntax is usually the least of my grievances in a new language. It usually has to do with features missing from other languages.
Lately, I've been doing javascript a lot more, and at first, I didn't like typing "function() {" all the time (and still don't). But now, I find that javascript doesn't have method_missing (only FF implements non-standard __noSuchMethod__), nor can you override the subscript operator. Gah! So then a thing like typing "function() {" fades into the background.
Erlang's syntax seems a lot less weird if you're familiar with Prolog, FWIW. It was a originally a DSL built on top of SICSTus Prolog, but seems to have accumulated some extra syntax along the way. Same with a lot of its other quirks.
It's like with foreign languages, if you have to learn a new language but at least their pronunciation rules and alphabet are the same as the one that you already know how to use the barrier is a lot lower than if they use a different script as well.
It serves as a bridge from the known in to the unknown.
As soon as that bridge isn't there the number of people that will be able (or willing) to make the jump drops dramatically.
If this step-in-between will make the concepts behind Erlang easier to bring to the masses then more of them will eventually make the step to Erlang itself as well.
I'm pretty sure that the syntax of Erlang is one of its main stumbling blocks, it is very far off the beaten path (and imo ugly), which means an immediate shortage of people that can program in it.
Grafting a C like language on top of the Erlang VM should theoretically give you the same kind of stability and scalability without the drawback of having to fish in a pool with all of 5 programmers in it (and they'll be working for a telco somewhere anyway).