Which would take little time, and unless any of them indicate that the two people that left together were actively arguing, what could you learn from them that shed any light on the actual disappearance, apart from a vague idea how drunk the missing guy might have been.
But if you're trying to being responsible rather than sensationalist, you can't just imagine possible lurid explanations without accepting that they're astonishingly unlikely to be true.
Even if there had been drunken disagreements in the pub or on the way home, how many tens of thousands of times a week does that happen, and how often is there a death as a result?
Also, if you're trying to be responsible, you should present the evidence for a 'scuffle' having occurred before putting it forward as even a possible piece of evidence in favour of something nasty having happened.
You have to be careful of judging people's actions with the benefit of hindsight.
In the context of asking about events at the point where a person is only known to be missing, someone might well not think it was important to question people individually. Likewise, the police may be perfectly happy with a summary of what people knew from someone in the group who spoke decent French.
Later, when the guy still hasn't returned, unless they thought there was something that didn't ring true, they may see no reason to formally ask people the same questions again, even if they might well informally talk about events.
Indeed, people would be fairly unlikely not to talk about events, and if nothing odd popped up in the course of normal chatter, that might well satisfy people that nothing had happened rather better than further formal interviews would.
Bearing in mind the high probability of accident over foul play, you have to understand that questioning people excessively can effectively be saying that you suspect their mate of involvement in their disappearance, which is something best not done without justification.