A couple folks over on Artima have jumped on me for the "Perl is generally easier for newbs than Smalltalk" comment I made in Ovid's journal, which I was somewhat surprised at.
It seems a strange argument to use the number of sigils as the basis for the "ease" of a language. Your average Perl newb will probably only see about 3 - @, %, $, plus 2 special variables, @_ and $_.
What I guess James doesn't know, or doesn't realize, is that in the case of @, % and $, they are actually meant to *help* readability. If a newb sees @, he knows it's an array. If he sees %, he knows it's a hash, etc.
Unfortunately, I never completed my PhD thesis on "Computer Programming and the Effect of Sigils on the Human Psyche", so I can't say with total certainty that this helps Perl newbs along or not, but I suspect it does. I think it's only when you get into advanced Perl, and advanced data structures in Perl, that the syntax (and sigils) start to get overbearing.
Once I became more versed with Perl, the sigils started to get in the way, and that's one of the reasons why I do more Ruby these days than Perl. I started to realize that I was only using scalars the vast majority of the time anyway, and that scalar could be anything - a simple scalar or a quadruply nested hash reference. The sigils no longer helped, and I found myself using a GUI debugger just to understand the structure's layout. But, that's another story...