So, he's observing that there is a strong correlation between scientists drinking a lot of beer and publishing less / being cited less. Causation does not nece

So, he's observing that there is a strong correlation between scientists drinking a lot of beer and publishing less / being cited less. Causation does not necessarily go in either direction (spending a lot of time in bars could lead to less time in the lab, continual failure to get published could drive them to drink, or both, or there could be a third variable), but I can't say the results are surprising.