Interesting design, though it did have hardware increment and decrement and some bit manipulation instructions. The IBM 1620 also didn’t have addition / subtraction hardware. It was nicknamed CADET for ‘Can’t Add, Doesn’t Even Try’. It used 6-bit characters (BCD for numbers) with a variable width word length. For addition, the memory contained a table of the result of adding any decimal digit to any other, arranged such that you could look it up by concatenating the digits to get the address. The result was a 6-bit character containing the decimal value and (optionally) a carry bit. If you the carry bit was set, you had to add 1 to the result of the next character pair that you added together. Multiplication worked a similar way, with a table of the results of single-digit multiplications and a loop doing long multiplication in software.
I think it’s a bit inaccurate to describe the Nanoprocessor as a RISC; it’s an accumulator architecture.
There are 16 storage registers but they can’t be used for most operations.
I would call it a microcontroller that uses external storage.