The typing discipline of a programming language is not an "issue." It's an engineering trade-off. Languages are designed with specific goals in mind and there are advantages and disadvantages for static and dynamic typing.

Similarly for writing small scripts. Smalltalk was never designed as a scripting language.

No programming language is perfect. Every language has pros and cons. You must choose the language that is best-suited for the application you have in mind. There is no such thing as a universal programming language.

The big draw of Smalltalk is its live programming environment. This is made easy by the total control of the environment (image plus GUI). Combined with the language's simplicity and elegance and you have a killer tool for massive productivity. I explain this further here: https://richardeng.medium.com/there-are-always-trade-offs-you-must-choose-ba859c184a17.

There's no question that some programming languages are much more difficult to learn than others. C++ and Rust are much tougher than Python and Smalltalk. Syntactical complexity is a big deal and it always has been.

Finally, if you limit yourself to popular languages that have large ecosystems, you will never adopt new languages in the future. You will be stuck with the current crop of languages forever.

FORTRAN and COBOL once ruled the IT universe. Then C, C++, and Java rose to displace them. Later came Python and JavaScript.

The IT world is constantly changing and evolving. Twenty years ago, Python was an insignificant language. In the 2010s, it took off because of the rise of data science and a huge ecosystem developed.

The same thing can happen with Smalltalk. If people recognize the value of the language (low cognitive load, superior OOP implementation, high productivity, etc.), then it too can rise in popularity.