It's a unary numeral system , but it doesn't really have a base, because it's not a positional number system.
Another similar thing to think about is the Roman numeral system. it starts out as unary system, but quickly turns into a mix of base 5 and 10 and even without a fixed position. That's also not a positional number system even if can be considered to have a base and a secondary base..
I think it's reasonable to say it completes that pattern of basal numbers. Saying that it's not positional is like saying that the base10 number 5555 isn't positional - it's just that all the digits happen to be the same.
Anyway, 5555 is just one number in the decimal system that fulfills the requirement that the position of digits is irrelevant, whereas most decimal numbers do not. In the tally mark system all numbers fulfill this requirement.
However, the thing I like most about it is that you'll never need to prove that I+I=II. It literally is II.
I remember reading the book called "Gödel, Escher, Bach" which is about Gödels incompleteness theorem. At some point it comes across this kind of thing and demonstrates how any natural number is the successor of the previous number, basically defining numbers as tally marks. From there it goes on to demonstrate why math itself is incomplete. It's kinda a fat book, but if you're into numbers, logic and coding it's a must read.
Base 1 does make a huge difference though. In base 1, number length is linear with respect to the size of the number - i.e. the number 10 needs 10 tallies. For any other base, the number length is logarithmic, i.e. the number 10 needs just 2 numerals in base 10 and still only 4 in base 2.
This is actually important in theoretical computer science, since computers would be much slower if you don't assume an "efficient encoding" of numbers, i.e. a logarithmic sized encoding. Base 1 is not efficient.