Two quarters of negative growth has only ever been a rule-of-thumb that mostly matched, the official determination has been with the National Bureau of Economic Research for a long long time ( https://www.nber.org/ ).
You can see on this graph that the rule-of-thumb didn't hold right at the beginning of the graph in 1947 when no recession was declared, nor in 2008 when it started with a single quarter, or even 2001 when it started with a positive quarter: https://fred.stlouisfed.org/series/A191RL1Q225SBEA