Here's is my approach:
$a=q_1n+1$
$b=q_2n+1$
$ab = q_1q_2n^2+q_1n+q_2n+1=(q_1q_2n+q_1+q_2)n+1$
I'm not sure if this is sufficient, and if so, whether there's a better proof.
Here's is my approach:
$a=q_1n+1$
$b=q_2n+1$
$ab = q_1q_2n^2+q_1n+q_2n+1=(q_1q_2n+q_1+q_2)n+1$
I'm not sure if this is sufficient, and if so, whether there's a better proof.
Yes this is good. One could also use modular arithmetic to say that $a\equiv 1\pmod n$ and $b\equiv 1\pmod n$ gives $ab\equiv 1\cdot 1\equiv 1\pmod n$, however I like your approach better because it uses the actual remainder as in the division algorithm, rather than using the relationship between remainders and modular arithmetic.
$(ab-1)=\underbrace{(a-1)b}_\text{divisible by n}+\underbrace{(b-1)}_\text{divisible by n}$
But it is not much different from your proof :
rewrite it $(a-1)(b-1)+(a-1)+(b-1)$ and replace $(a-1)$ by $q_1n$ and same for $b$.